摘要
大数据聚类在无线传感网络数据处理领域中具有重要意义,但是大数据聚类方法存在聚类效果不佳、Jaccard系数较低等问题,提出基于粒子群算法的无线传感网络大数据优化方法。该方法结合主成分分析方法和信息熵降维处理大数据,减少数据聚类所需的时间,采用直觉模糊核聚类算法聚类大数据,引入粒子群算法,优化直觉模糊核聚类方法,利用优化后的算法获得无线传感网络大数据聚类的优化结果,实现大数据聚类。仿真分析结果表明,所提方法的聚类效果较好,Jaccard系数在0.70以上,数据平均熵仅为0.36,并且时间复杂度仅为26.3%,该方法的应用价值更高。
Big data clustering is of great significance in the field of wireless sensor network data processing,but the big data clustering method has problems of poor clustering effect and low Jaccard coefficient.A particle swarm optimization method for wireless sensor net⁃work big data is proposed.Principal component analysis and information entropy dimensionality reduction is combined to process big da⁃ta,the time required for data clustering is reduced intuitionistic fuzzy kernel clustering algorithm is adopted to cluster big data,particle swarm optimization algorithm is introduced,intuitionistic fuzzy kernel clustering method is optimized,the optimized algorithm is used to obtain the optimization results of wireless sensor network big data clustering,and big data clustering is realized.The simulation results show that the clustering effect of the proposed method is good,the Jaccard coefficient is above 0.70,the average entropy of data is only 0.36,and the time complexity is only 26.3%.This method has higher application value.
作者
程宁
李超
CHENG Ning;LI Chao(School of Information Engineering,Hubei Light Industry Technology Institute,Wuhan Hubei 430070,China;Information Construction and Management Division,Hubei University,Wuhan Hubei 430062,China)
出处
《传感技术学报》
CAS
CSCD
北大核心
2023年第8期1316-1322,共7页
Chinese Journal of Sensors and Actuators
基金
教育部科技发展中心产学研创新基金新一代信息技术创新项目(2018A03021)。
关键词
无线传感网络
大数据聚类
粒子群算法
主成分分析
信息熵
直觉模糊核聚类算法
wireless sensor network
big data clustering
particle swarm optimization
principal component analysis
information entropy
intuitionistic fuzzy kernel clustering algorithm
