摘要
聚类分析通过数据挖掘、机器学习技术,进行数据查找,传统聚类分析中串行方法的数据处理,要求处理内存大数据,CPU运算速度快,无法进行海量数据分析。高斯混合模型中概率密度函数,精确地处理海量数据,将其分配到各个混合成分中,实现精细化查找,并简化数据处理步骤。高斯混合模型利用最大期望(EM)算法进行参数评估,显著提高数据分析速度。基于最大期望聚类算法原理,对高斯混合模型进行优化,旨在实现海量数据的准确运算。高斯模型结合Hadoop平台中的海量数据进行算法分析,对大数据的可视化处理和样本分析提供帮助。研究结果显示:最大期望聚类算法可以进行海量数据分析,通过简化运算步骤,实现短时间内的数据查找、分析和处理。
Clustering analysis uses data mining and machine learning techniques to search data,the traditional clustering analysis method uses serial method to process data,it has high requirements on computer memory and CPU,it is not suitable for mass data analysis.Gaussian mixture model uses the probability density function to accurately quantify the data,distribute the data to the various mixed components,in order to find the refinement of the data,simplify the data analysis steps.At the same time,the Gaussian mixture model is used to analyze the data,the maximum expectation（EM）algorithm is mainly used to evaluate the parameters to further,they improve the data analysis efficiency.This paper mainly introduces the principle of maximum expectation clustering algorithm and optimizes the Gaussian mixture model,the purpose is to make it adapt to a wider range.At the same time,combined with the case data on the Hadoop platform,the algorithm is analyzed and the probability density of the sample analysis is displayed by using the visual graph.The study finds that the maximum expectation clustering algorithm is suitable for mass data analysis,which is calculated by Gaussian mixture model,greatly improving the efficiency of clustering analysis.
作者
何庆
易娜
汪新勇
江立斌
He Qing;Yi Na;Wang Xinyong;Jiang Libin(China Mobile Group Southern Base, Guangzhou 510623)
出处
《微型电脑应用》
2018年第5期50-52,75,共4页
Microcomputer Applications
关键词
高斯混合模型
最大期望
聚类算法
Gaussian mixture model
Maximum expectation
Clustering algorithms
