时间序列模糊关联规则的挖掘

Fuzzy Assosiation Rules Mining from Time Series

对于许多复杂系统产生的时间序列,研究序列的局部行为和局部关联特征往往比原来的研究系统全局性模型具有明显的优势。为研究时间序列内部或时间序列间局部形态的关联特征,文章借助模糊集来软化时间序列属性论域的划分边界从而研究时间序列局部形态的模糊关联规则、规则可信度和规则的评价方法。实际算例显示了算法的有效性。

On the occasion of dealing with time series hailed from complex system,the investigation of series's local patterns and local relationship has distinct superiority over traditional global models.In order to find rules relating pat-terns in a time series to other patterns in that series,or patterns in one series to patterns in another series,a fuzzy sub-series discretization method,which soften the effect of sharp boundaries of delegate of each local sub-series,is pro-posed.Then the local patterns's relationship called fuzzy association rules,rules's confidence and rules's selection mea-sure are studied.The practical calculation shows that the mining of fuzzy assosiation rules is more effective.