摘要
相似性度量在聚类算法设计中起关键作用,使用合适的距离度量函数能够反映数据对象间的相似性。本文对聚类算法中数据对象间相似性度量的特征进行了系统性归纳总结,通过MapReduce编程模型实现对各种相似性度量聚类算法的实验比较分析,将为聚类分析研究者提供参考。
The similarity measure plays a key role in clustering algorithms. Using appropriate distance measure function can reflect the similarity between data objects. This paper aims to conduct a systematic summary on data objects similarity measure in clustering algorithms. The paper will also implement comparative analysis on various similarity measure clustering algorithms by MapReduce programming model,which can provide references to researchers on clustering algorithms.
作者
彭天昊
潘有顺
杨胜林
PENG Tianhao;PAN Youshun;YANG Shenglin(Moutai Institute,Department of Brewing Engineering Automation,Renhuai 564507,China)
出处
《现代信息科技》
2018年第11期10-12,共3页
Modern Information Technology