云计算设备中的大数据特征高效分类挖掘方法研究

摘要

The big-data classification mining in cloud computing equipment is the basis of real pattern recognition and intel-ligent control. The topological-structure grid-partition mining algorithm is adopted for large data mining in traditional cloud com-puting equipment,which can not effectively extract the detail characteristics of big data due to its poor classifying veracity. A ef-ficient classification mining algorithm for big data feature of cloud computing equipment is proposed,which is based on the frac-tional Fourier transform feature matching and K-L classification. The big data storage mechanism system of cloud computing equipment is analyzed. The fractional Fourier transform is used to deal with feature extraction and feature matching of big data. On the basis of K-L transform,the optimal path is chosen to guide categorical space,and a K-L big-data feature classifier is es-tablished to realize classification mining in cloud computing equipment. The simulation results show that this algorithm has ad-vantages of high-accuracy feature classification mining,less energy consumption and higher efficiency,and can realize efficient classification mining for the big data feature of the cloud computing equipments.%云计算设备中的大数据分类挖掘是现实模式识别和智能控制的基础,传统方法中对云计算设备中的大数据挖掘采用拓扑结构网格分区挖掘算法,不能有效提取大数据的细节特征,分类的准确性不好.提出一种基于分数阶Fourier变换特征匹配和K-L分类的云计算设备中的大数据特征高效分类挖掘算法.进行云计算设备中大数据存储机制体系分析,采用分数阶Fourier变换进行云计算设备中大数据特征提取和大数据特征匹配,基于K-L变换,选择最优的路径进行分类空间导引,构建了K-L大数据特征分类器,进行云计算设备中的大数据特征分类挖掘.仿真结果表明,采用该算法进行云计算设备中的大数据特征分类挖掘,特征分类挖掘的准确度较高,能量开销较少,效率较高.