关于分布式数据库准确分类仿真研究
合集下载
相关主题
- 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。
Research on Accurate Classification and Simulation of Distributed Databases
CAO Man-man1 ,WANG Mian2
(1. Department of Computer Science, Jining University, Qufu Shandong 273155 , China; 2. Institute of Scientific and Technical Information of Jining, Jining Shandong 272000, China)
第36卷第1期 文章编号:1006-9348 (2019)Fra Baidu bibliotek01 -0354 -04
计算机仿真
2019年1月
关于分布式数据库准确分类仿真研究
曹曼曼-汪勉2
(1.济宁学院计算机科学系,山东曲阜273155 ;2.济宁市科技情报研究所,山东济宁272000)
摘要:对分布式数据库进行准确分类,能够有效提高数据的利用率。对数据库的准确分类,需通过近似函数计算后验概率, 根据概率结果,完成数据库的准确分类。传统方法通过构造查询矩阵和相似度矩阵,确定数据库准确分类的策略,但忽略了 后验概率的计算,导致分类效果不显著。在云计算平台下,提出基于Parzen窗估计模型的分布式数据库准确分类方法,在分 析分布式数据库分类系统原理模型基础上,利用Parpen窗估计模型确定分布式数据库区间样本的类别条件概率密度函数, 通过插值法设计类别条件概率密度函数的近似函数,并利用此近似函数计算数据库分类样本后验概率,根据概率结果,实现 分布式数据库分类。通过计算数据库分类结果的亲和力,并将分类结果亲和力与设定阈值进行对比,实现分布式数据库准 确分类。实验结果表明,所提方法分类准确度较高,且分类过程较简单。 关键词:云计算平台;分布式;数据库;准确分类 中图分类号:TP311 文献标识码:B
1引言
分布式数据库是指分散的多个数据存储单元连接起来 组成一个逻辑上统一的数据库。面对数据量的井喷式增长 和不断增长的用户需求,分布式数据库如何准确分类得到相 关专家学者的重视⑴。对云计算平台下分布式数据库进行 分类,能够使数据准确的存储到数据库中,为后续数据处理 提供方便条件"却。为保证云计算平台下分布式数据库分类
ABSTRACT: To accurately classify distributed database can effectively improve the utilization rate of data. The tradi tional method constructs the query matrix and similarity matrix and determines the strategy of accurate classification for database, but ignores the calculation of posterior probability, which results in the insignificant classification effect. In cloud computing platform, this article puts forward an accurate classification method of distributed database based on Parzen window estimation model. Based on the analysis of the principle model of classification system in distribu ted database, this research used Parzen window estimation model to determine the probability density function of class condition of interval sample in distributed database. Then, our research used the interpolation method to design the approximate function of probability density function of class condition and used this approximate function to calculate the posterior probability of classification sample in database. According to the probability result, the research realized the classification of distributed database. By calculating the affinity of database classification result and comparing the affinity of classification result with the set threshold, we achieved the accurate classification of distributed database. Simulation results show that the proposed method has high classification accuracy and simple classification process. KEYWORDS: Cloud computing platform ; Distributed ; Database ; Accurate classification