【摘要】 在云计算环境下多来源数据最优选取过程中,来自不同的服务器数据具有不同特征,数据所含信息的融合度也不同,使得数据可选取特征存在较强的关联性.传统依的数据最优选取模型,在进行选择时,这种关联性会给最优数据的选择带来干扰,选出大量的相关数据,不能很好的排除关联性数据的排除,无法实现数据的最优高效选取,提出一种依据数据估计量变化的云计算环境下多来源数据最优选取模型,利用前几点数据拟合直线或曲线,进行云计算环境下多来源数据最优选取中干扰数据的预测,分析干扰数据点删除前后推断统计量的变化,检测该干扰数据点是不是强干扰点,过滤该强干扰数据点后的模型则是云计算环境下多来源数据最优选取模型,通过分析数据排除模型引起的估计量的变化,实现云计算环境下多来源数据的最优选取。仿真实验结果说明,在大量云计算环境下多来源数据的处理中,采用所提模型可有效选取最优数据,并且具有较高的准确率。更多还原

【Abstract】 Data from multiple sources in cloud computing environment,the optimal selection process,the data from different servers have different characteristics,the data contained information of alignment is also different,the data can be selected characteristics exist strong correlation.According to the traditional optimal selection model of data for when the choice,this kind of correlation interference to the optimal choice of data,select a lot of related data,cannot very good elimination correlation data,choosing the best efficient unable to implement the data,puts forward a kind of according to change data estimator under the cloud computing environment of multiple sources of data,the optimal selection model,the use of some data before fitting a straight line or curve,data from multiple sources in cloud computing environment data to predict the optimal selection of interference,interference analysis concluded that the change of the statistic,before and after the data point to delete test data points are strong interference,the interference model after filtering the noise data points is a cloud computing environment from multiple sources data optimal selection model,through the analysis of data to eliminate model caused by the change of the estimator,realize the cloud computing environment,the optimal selection of data from multiple sources.Simulation experiment results show that,in a large number of data from multiple sources in cloud computing environment,the proposed model can effectively choose the optimal data,and has high accuracy.更多还原