An incremental mixed data clustering method using a new distance measure

Fakhroddin Noorbehbahani, Seyed Rasoul Mousavi, Abdolreza Mirzaei

Research output: Contribution to journalArticle

11 Citations (Scopus)
1 Downloads (Pure)

Abstract

Clustering is one of the most applied unsupervised machine learning tasks. Although there exist several clustering algorithms for numeric data, more sophisticated clustering algorithms to address mixed data (numeric and categorical data) more efficiently are still required. Other important issues to be considered in clustering are incremental learning and generating a sufficient number of clusters without specifying the number of clusters a priori. In this paper, we introduce a mixed data clustering method which is incremental and generates a sufficient number of clusters automatically. The proposed method is based on the Adjusted SelfOrganizing Incremental Neural Network (ASOINN) algorithm exploiting a new distance measure and new update rules for handling mixed data. The proposed clustering method is compared with the ASOINN and three other clustering algorithms comprehensively. The results of comparative experiments on various data sets using several clustering evaluation measures show the effectiveness of the proposed mixed data clustering method.
Original languageEnglish
Pages (from-to)731-743
Number of pages13
JournalSoft Computing
Volume19
Issue number3
Early online date6 May 2014
DOIs
Publication statusPublished - Mar 2015

Fingerprint

Mixed Data
Data Clustering
Distance Measure
Clustering Methods
Clustering algorithms
Number of Clusters
Clustering Algorithm
Clustering
Self-organizing
Numerics
Neural networks
Neural Networks
Sufficient
Incremental Learning
Learning systems
Nominal or categorical data
Unsupervised Learning
Network Algorithms
Machine Learning
Update

Keywords

  • Mixed data clustering
  • Incremental learning
  • Distance measure
  • Clustering evaluation measures
  • SOM

Cite this

An incremental mixed data clustering method using a new distance measure. / Noorbehbahani, Fakhroddin ; Mousavi, Seyed Rasoul; Mirzaei, Abdolreza.

In: Soft Computing, Vol. 19, No. 3, 03.2015, p. 731-743.

Research output: Contribution to journalArticle

Noorbehbahani, Fakhroddin ; Mousavi, Seyed Rasoul ; Mirzaei, Abdolreza. / An incremental mixed data clustering method using a new distance measure. In: Soft Computing. 2015 ; Vol. 19, No. 3. pp. 731-743.
@article{0bf3d424f6904accb08c5d14320136cd,
title = "An incremental mixed data clustering method using a new distance measure",
abstract = "Clustering is one of the most applied unsupervised machine learning tasks. Although there exist several clustering algorithms for numeric data, more sophisticated clustering algorithms to address mixed data (numeric and categorical data) more efficiently are still required. Other important issues to be considered in clustering are incremental learning and generating a sufficient number of clusters without specifying the number of clusters a priori. In this paper, we introduce a mixed data clustering method which is incremental and generates a sufficient number of clusters automatically. The proposed method is based on the Adjusted SelfOrganizing Incremental Neural Network (ASOINN) algorithm exploiting a new distance measure and new update rules for handling mixed data. The proposed clustering method is compared with the ASOINN and three other clustering algorithms comprehensively. The results of comparative experiments on various data sets using several clustering evaluation measures show the effectiveness of the proposed mixed data clustering method.",
keywords = "Mixed data clustering, Incremental learning, Distance measure, Clustering evaluation measures, SOM",
author = "Fakhroddin Noorbehbahani and Mousavi, {Seyed Rasoul} and Abdolreza Mirzaei",
year = "2015",
month = "3",
doi = "10.1007/s00500-014-1296-7",
language = "English",
volume = "19",
pages = "731--743",
journal = "Soft Computing",
issn = "1432-7643",
publisher = "Springer Verlag",
number = "3",

}

TY - JOUR

T1 - An incremental mixed data clustering method using a new distance measure

AU - Noorbehbahani, Fakhroddin

AU - Mousavi, Seyed Rasoul

AU - Mirzaei, Abdolreza

PY - 2015/3

Y1 - 2015/3

N2 - Clustering is one of the most applied unsupervised machine learning tasks. Although there exist several clustering algorithms for numeric data, more sophisticated clustering algorithms to address mixed data (numeric and categorical data) more efficiently are still required. Other important issues to be considered in clustering are incremental learning and generating a sufficient number of clusters without specifying the number of clusters a priori. In this paper, we introduce a mixed data clustering method which is incremental and generates a sufficient number of clusters automatically. The proposed method is based on the Adjusted SelfOrganizing Incremental Neural Network (ASOINN) algorithm exploiting a new distance measure and new update rules for handling mixed data. The proposed clustering method is compared with the ASOINN and three other clustering algorithms comprehensively. The results of comparative experiments on various data sets using several clustering evaluation measures show the effectiveness of the proposed mixed data clustering method.

AB - Clustering is one of the most applied unsupervised machine learning tasks. Although there exist several clustering algorithms for numeric data, more sophisticated clustering algorithms to address mixed data (numeric and categorical data) more efficiently are still required. Other important issues to be considered in clustering are incremental learning and generating a sufficient number of clusters without specifying the number of clusters a priori. In this paper, we introduce a mixed data clustering method which is incremental and generates a sufficient number of clusters automatically. The proposed method is based on the Adjusted SelfOrganizing Incremental Neural Network (ASOINN) algorithm exploiting a new distance measure and new update rules for handling mixed data. The proposed clustering method is compared with the ASOINN and three other clustering algorithms comprehensively. The results of comparative experiments on various data sets using several clustering evaluation measures show the effectiveness of the proposed mixed data clustering method.

KW - Mixed data clustering

KW - Incremental learning

KW - Distance measure

KW - Clustering evaluation measures

KW - SOM

U2 - 10.1007/s00500-014-1296-7

DO - 10.1007/s00500-014-1296-7

M3 - Article

VL - 19

SP - 731

EP - 743

JO - Soft Computing

JF - Soft Computing

SN - 1432-7643

IS - 3

ER -