arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2306.09256cs.LGcs.DB
收藏

若干基于密度的聚类技术综述

A Survey of Some Density Based Clustering Techniques

Icfai University Nagaland(那兰市伊克法伊大学);Sikkim Manipal Institute of Technology(锡金马尼帕尔理工学院)

Rupanka Bhuyan, Samarjeet Borah

AI总结:

本文对DBSCAN、OPTICS等多种基于密度的聚类方法进行综述,分析其特点、优缺点及在不同数据集上的适用性,为数据挖掘中的模式提取提供方法参考。

摘要:

基于密度的聚类是数据挖掘中一类用于从数据集中提取未知模式的聚类方法。目前已有多种基于密度的聚类方法,例如DBSCAN、OPTICS、DENCLUE、VDBSCAN、DVBSCAN、DBCLASD和ST-DBSCAN。本文对这些方法展开了研究,分析了它们的特点、优缺点,重点探讨了它们在不同类型数据集上挖掘有用且合适模式的适用性。

英文摘要:

Density Based Clustering are a type of Clustering methods using in data mining for extracting previously unknown patterns from data sets. There are a number of density based clustering methods such as DBSCAN, OPTICS, DENCLUE, VDBSCAN, DVBSCAN, DBCLASD and ST-DBSCAN. In this paper, a study of these methods is done along with their characteristics, advantages and disadvantages and most importantly, their applicability to different types of data sets to mine useful and appropriate patterns.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑