摘要
关联规则约束下云服务组合大数据挖掘容易出现邻阶干扰,数据挖掘的聚类性和抗干扰性不佳。为提高云服务组合大数据挖掘能力,提出一种基于模糊卷积神经网络的大数据分类挖掘技术。采用连续模板匹配技术进行大数据的分布式数据结构分析,结合匹配相关检测方法进行云服务组合大数据的信息融合处理,对高维融合数据进行频繁项检测和关联规则特征提取;对提取的云服务组合大数据的关联规则采用模糊卷积神经网络分类器进行属性分类,结合特征压缩方法对分类输出的云服务组合大数据进行降维处理;采用模糊聚类方法实现对云服务组合大数据的分类挖掘。仿真结果表明:采用该方法进行云服务组合大数据挖掘的聚类性能较好,在挖掘精度和抗干扰性能表现方面具有优势。
Under the constraint of association rules,big data mining cloud service composition is prone to appear neighbor interference,and the clustering and anti-interference of data mining is not good.This paper presents a classification mining technique for big data based on fuzzy convolution neural network.Using continuous template matching technology to analyze big data’s distributed data structure,and combining matching correlation detection method to deal with the information fusion of cloud service assemblage big data the frequent item detection and association rule feature extraction are carried out for high-dimensional fusion data,and fuzzy convolution neural network classifier is used to classify the extracted association rules of cloud service combination big data.The feature compression method is used to reduce the dimension of the cloud service composition big data,and the fuzzy clustering method is used to realize the classification mining of the cloud service composition big data.The simulation results show that the clustering performance of this method for cloud service composition big data mining is good,and the mining accuracy and anti-interference performance are superior.
作者
林倩瑜
LIN Qianyu(Jimei University Cheng Yi College,Xiamen 361021,China)
出处
《重庆理工大学学报（自然科学）》
CAS
北大核心
2019年第10期121-126,共6页
Journal of Chongqing University of Technology：Natural Science
基金
福建省青年教师教育科研项目“基于大数据的网络舆情爬虫算法的研究与设计”
关键词
模糊卷积神经网络
大数据
挖掘
分类
特征提取
fuzzy convolution neural network
big data
mining
classification
feature extraction
