文献类型：期刊文献

中文题名：海量数据干扰下冗余数据高性能消除方法

英文题名：High performance elimination method for redundant data under massive data interference

作者：茹蓓[1];李虹[2]

第一作者：茹蓓

机构：[1]新乡学院计算机与信息工程学院;[2]吉林省质监检测基地吉林省纤维检验处

第一机构：新乡学院计算机与信息工程学院

年份：2017

卷号：39

期号：6

起止页码：686-690

中文期刊名：沈阳工业大学学报

外文期刊名：Journal of Shenyang University of Technology

收录：CSTPCD;;Scopus;北大核心:【北大核心2014】；

基金：河南省科技厅科技攻关资助项目(172102210445);河南省科技厅软科学研究资助项目(152400410345);河南省教育厅资助项目(15A520093)

语种：中文

中文关键词：海量数据;干扰;冗余数据;高性能;消除方法;改进;均值;传递函数

外文关键词：massive data;;interference;;redundant data;;high performance;;elimination method;;improvement;;mean value;;transfer function

摘要：针对海量数据处理过程中大量相似特征会给数据分类造成冗余干扰,在分类中心确定时出现多次校验、重复等弊端,提出一种海量数据干扰下冗余数据高性能消除方法.采用主动采样方法提取海量数据干扰下冗余数据特征,并对其进行分类.引入均值漂移传递函数对冗余数据进行分类处理,获取冗余数据活跃程度,实现冗余数据的高性能消除.结果表明,相比传统的消除方法,高效消除方法性能良好,所需时间短,具有一定的优越性.
Aiming at the problem that in the data treatment process of massive data,a lot of similar characteristics can bring the redundant interference to the data classification,and cause such drawbacks as multiple check and repetition during the determination with the classification center,a high performance elimination method for redundant data under the massive data interference was proposed. With the active sampling method,the redundancy data feature under the massive data interference was extracted and classified. In addition,the mean shift transfer function was introduced to perform the classification of redundant data, obtain the redundant data activity and realize the high performance elimination of redundancy data. The results showthat compared with the traditional method,the high performance elimination method has good properties,short processed time and certain superiority.