Looking for natural patterns in data - Part 1. Density-based approach

被引：202

作者：

Daszykowski, M ^{[1
]}

Walczak, B ^{[1
]}

Massart, DL ^{[1
]}

机构：

[1] FABI VUB, ChemoAC, B-1090 Brussels, Belgium

来源：

CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS | 2001年 / 56卷 / 02期

关键词：

pattern recognition; density-based clustering; outliers and inliers identification;

D O I：

10.1016/S0169-7439(01)00111-3

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

A density-based unsupervised clustering approach for detecting natural patterns in data (further denoted as NP) is presented, and its performance is illustrated for data sets with different types of clusters. NP works for arbitrary clusters, is a single-scan technique, requires no presumptions regarding data distribution and requires only one input parameter, which describes the minimal number of objects, considered as cluster. Moreover, a comparison of NP with partitioning approaches is demonstrated. NP can be applied not only for data clustering, but also for the identification of outliers. (C) 2001 Elsevier Science B.V. All rights reserved.

引用

页码：83 / 92

页数：10

共 14 条

[1]

BISHOP CM, 1996, ADV NEURAL INFORMATI, V8, P456

[2] ART-2 - SELF-ORGANIZATION OF STABLE CATEGORY RECOGNITION CODES FOR ANALOG INPUT PATTERNS [J].