大数据环境下并行数据流预测及优化方法

Analysis of integrated online association rules based on OLAP fuzzy data cube

以模糊集成联机分析处理(On-Line Analytical Processing,OLAP)为基础对关联规则挖掘算法进行改进,该算法为多维联机研究提供一种关联规则计算架构。基于模糊数据立方体的知识发现,为用户提供灵活的多维数据层次抽象模式。在多维数据集的多维属性处理中,引入模糊数据立方体作为问题措施补充,并利用不同层次的模糊关联规则构造模糊数据立方体,然后利用权重和多层次的概念构建模糊加权多层次关联规则。最后,通过对所提算法在合成数据集以及2000年中国人口普查的数据仿真测试,验证了基于OLAP的挖掘方法要比离散关联规则挖掘方法、单独支持阈值关联规则及最小挖掘项集关联规则三种对比算法,在最小支持度、置信度、权重均值等指标上,性能更加优异。

Based on the fuzzy integrated online analytical processing(OLAP),the algorithm of association rule mining is improved.This algorithm provides a new association rule mining method for multi dimensional online research.Knowledge discovery based on fuzzy data cube provides a flexible and multi dimensional data level abstraction model for users.The fuzzy data cube is used as the supplementary measures for problem to deal with the multidimensional attributes of the multidimensional data set.Then,the fuzzy data cube is constructed by using the fuzzy association rules of different levels,and the fuzzy weighted multilevel association rules are constructed by using the weights and multiple levels.Finally,through the simulation test of the proposed algorithm in the synthetic data set and the 2000 China census data show that,the OLAP based mining method was more excellent in the minimum support degree,upper confidence degree and weight average index than the three comparison algorithm of discrete association rules mining method,support a separate threshold association rules and mining the smallest association rule set.