最近最远得分的聚类性能评价指标

PDF下载HTML

Title:

A clustering evaluation index based on the nearest and furthest score

作者:

冯柳伟^1,2, 常冬霞^1,2, 邓勇³, 赵耀^1,2

1. 北京交通大学 信息科学研究所, 北京 100044;
2. 北京交通大学 计算机与信息科学学院, 北京 100044;
3. 中国科学院 软件研究所, 北京 100190

Author(s):

FENG Liuwei^1,2, CHANG Dongxia^1,2, DENG Yong³, ZHAO Yao^1,2

1. Institute of Information Science, Beijing Jiaotong University Beijing 100044, China;
2. School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China;
3. Institute of Software, Chinese Academy of Sciences, Beijing 100190, China

关键词:

最近邻一致性; 最远邻相异性; K-means聚类算法; 评分机制; 评价指标; 层次聚类

Keywords:

the nearest neighbor consistency; the furthest neighbor difference; K-means clustering algorithm; scoring mechanism; evaluation index; hierarchical clustering

分类号:

TP391

DOI:

10.11992/tis.201611007

摘要:

聚类算法是数据分析中广泛使用的方法之一，而类别数往往是决定聚类算法性能的关键。目前，大部分聚类算法需要预先给定类别数，在很多情况下，很难根据数据集的先验知识获得有效的类别数。因此，为了获得数据集的类别数，本文基于最近邻一致性和最远邻相异性的准则，提出了一种最近最远得分评价指标，并在此基础上提出了一种自动确定类别数的聚类算法。实验结果证明了所提评价指标在确定类别数时的有效性和可行性。

Abstract:

The clustering algorithm is one of the widely-used methods in data analysis. However, the number of clusters is essential to determine the performance of the clustering algorithm. At present, the number of clusters usually need to be specified in advance. In most cases, it is difficult to obtain the valid cluster number according to a priori knowledge of the dataset. To obtain the number of clusters automatically, a Nearest and Furthest Score (NFS) index was proposed based on the principles of the nearest neighbor consistency and the furthest neighbor difference. Moreover, an Automatic Clustering NFS (ACNFS) algorithm was also proposed, which can determine the number of clusters automatically. The experimental results prove the index is reasonable and practicable to determine the cluster number.