摘要
针对物联网环境下实时数据挖掘中资源约束的特点,分析了快速有效地挖掘抽取知识的方法。在K-means算法的基础上,结合RA-Cluster算法,提出了ARRA-Kmeans聚类算法。并基于环境资源约束和时间约束的特点,引入自适应聚类方法和滑动窗口技术,对网络节点的实时数据进行挖掘。实验结果表明,随着流数据量的增大,ARRA-Kmeans算法在处理实时动态的数据时具有较好的效果,聚类精度较高,处理时间较快。
According to features of resource constraints in real time data mining on the Internet of Things, a method of fast and effective knowledge mining and extraction is analyzed. Combined with RA- Cluster algorithm ,the ARRA-Kmeans clustering algorithm is proposed on the basis of K-means algorithm. On the basis of the characteristics of environmental resource constraints and time constraints, the adaptive clustering method and the sliding window technique are introduced, and the real-time data of the network node are mined. The test results show that with the increasing amount of data flow, ARRA-Kmeans algorithm has good effect in the treatment of real-time dynamic data with good clustering accuracy and fast processing time.
出处
《北京信息科技大学学报(自然科学版)》
2014年第3期25-27,37,共4页
Journal of Beijing Information Science and Technology University
基金
"十二五"国家密码发展基金密码理论课题(MMJJ201101025)
关键词
资源约束
聚类
自适应
实时
滑动窗口
resource constraint
clustering
adaptive
real-time
sliding window