Comparing Clusterings - An Overview
Comparing Clusterings - An Overview
复制标题
DOI：
10.5445/ir/1000011477
复制
发表时间：
2007
期刊：
影响因子：
--
通讯作者：
Silke Wagner;D. Wagner
中科院分区：
文献类型：
--
作者：
Silke Wagner;D. Wagner
随着我们现在必须处理的数据量变得越来越大，帮助我们检测数据中的结构并识别数据中感兴趣的子集的方法变得越来越重要。这些方法之一是聚类，即将一组元素分割成子集，使得每个子集中的元素在某种程度上彼此“相似”，而不同子集的元素是“不相似”的。在文献中，我们可以找到各种各样的聚类算法，每一种都有一定的优点，但也有一定的缺点。在这方面出现的典型问题包括：
As the amount of data we nowadays have to deal with becomes larger and larger, the methods that help us to detect structures in the data and to identify interesting subsets in the data become more and more important. One of these methods is clustering, i.e. segmenting a set of elements into subsets such that the elements in each subset are somehow ”similiar” to each other and elements of different subsets are ”unsimilar”. In the literature we can find a large variety of clustering algorithms, each having certain advantages but also certain drawbacks. Typical questions that arise in this context comprise: