数据空间中一种基于步进的后映射(SPM)迭代方法  

An iterative method of step-by-step post-mapping (SPM) in data space

作　　者：吉露 曹斌 JI Lu;CAO Bin(College of Big Data and Information Engineering,Guizhou University,Guiyang 550025,China;Chinalco Intelligent Technology Development Co.Ltd.,Hangzhou 310000,China)

机构地区：[1]贵州大学大数据与信息工程学院,贵阳550025 [2]中铝智能科技发展有限公司,杭州310000

出　　处：《智能计算机与应用》2021年第9期22-27,共6页Intelligent Computer and Applications

摘　　要：数据空间是一个自我维持和自我管理的系统,需要支持中介模式及其模式映射之间的不确定性。通过将概率和可靠度分配给模式映射,有些映射的结果并不完全正确,这将不利于用户的精确查询请求。本文提出了一种SPM(step post-mapping)映射方法,利用可能性理论对可靠映射进行管理,将可靠映射集划分为可能性分布的子集,并给每个子集分配一个递归的可能性函数。使用铝电解真实数据集,通过查准率、查全率和F值评估算法,并将其与现有的方法进行对比分析,验证该方法的可行性。The data space is a self-sustaining and self-managing system. The uncertainty between mediation patterns and their schema mappings needs to be supported. By assigning probability and reliability to pattern mappings,some of the mappings are not exactly correct. This is harmful for the user’s accurate query requests. It is necessary to increase the probability and reliability of the mapping. In this paper,a SPM(Step Post-Mappin)mapping method is proposed to manage the reliable mapping using possibility theory. The set of reliable mappings is divided into subsets of probability distribution,and a recursive probability function is assigned to each subset. The actual dataset of aluminum electrolysis is used to compare and analyze the existing methods through the accuracy,recall and F value to verify the feasibility of the method.

关 键 词：数据空间 可靠度 模式映射 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]