A Comparative Study on TIBA Imputation Methods in FCMdd-Based Linear Clustering with Relational Data

A Comparative Study on TIBA Imputation Methods in FCMdd-Based Linear Clustering with Relational Data

Relational fuzzy clustering has been developed for extracting intrinsic cluster structures of relational data and was extended to a linear fuzzy clustering model based on Fuzzy c-Medoids (FCMdd) concept, in which Fuzzy c-Means-(FCM-) like iterative algorithm was performed by defining linear cluster...

Full description

Bibliographic Details
Main Authors: Takeshi Yamamoto, Katsuhiro Honda, Akira Notsu, Hidetomo Ichihashi
Format: Article
Language:English
Published: Hindawi Limited 2011-01-01
Series:Advances in Fuzzy Systems
Online Access:http://dx.doi.org/10.1155/2011/265170
id doaj-e730e87545184e49ac2b972117fb1576
record_format Article
spelling doaj-e730e87545184e49ac2b972117fb15762020-11-24T23:47:15ZengHindawi LimitedAdvances in Fuzzy Systems1687-71011687-711X2011-01-01201110.1155/2011/265170265170A Comparative Study on TIBA Imputation Methods in FCMdd-Based Linear Clustering with Relational DataTakeshi Yamamoto0Katsuhiro Honda1Akira Notsu2Hidetomo Ichihashi3Department of Computer Science and Intelligent Systems, Osaka Prefecture University, Osaka 599-8531, JapanDepartment of Computer Science and Intelligent Systems, Osaka Prefecture University, Osaka 599-8531, JapanDepartment of Computer Science and Intelligent Systems, Osaka Prefecture University, Osaka 599-8531, JapanDepartment of Computer Science and Intelligent Systems, Osaka Prefecture University, Osaka 599-8531, JapanRelational fuzzy clustering has been developed for extracting intrinsic cluster structures of relational data and was extended to a linear fuzzy clustering model based on Fuzzy c-Medoids (FCMdd) concept, in which Fuzzy c-Means-(FCM-) like iterative algorithm was performed by defining linear cluster prototypes using two representative medoids for each line prototype. In this paper, the FCMdd-type linear clustering model is further modified in order to handle incomplete data including missing values, and the applicability of several imputation methods is compared. In several numerical experiments, it is demonstrated that some pre-imputation strategies contribute to properly selecting representative medoids of each cluster.http://dx.doi.org/10.1155/2011/265170
collection DOAJ
language English
format Article
sources DOAJ
author Takeshi Yamamoto
Katsuhiro Honda
Akira Notsu
Hidetomo Ichihashi
spellingShingle Takeshi Yamamoto
Katsuhiro Honda
Akira Notsu
Hidetomo Ichihashi
A Comparative Study on TIBA Imputation Methods in FCMdd-Based Linear Clustering with Relational Data
Advances in Fuzzy Systems
author_facet Takeshi Yamamoto
Katsuhiro Honda
Akira Notsu
Hidetomo Ichihashi
author_sort Takeshi Yamamoto
title A Comparative Study on TIBA Imputation Methods in FCMdd-Based Linear Clustering with Relational Data
title_short A Comparative Study on TIBA Imputation Methods in FCMdd-Based Linear Clustering with Relational Data
title_full A Comparative Study on TIBA Imputation Methods in FCMdd-Based Linear Clustering with Relational Data
title_fullStr A Comparative Study on TIBA Imputation Methods in FCMdd-Based Linear Clustering with Relational Data
title_full_unstemmed A Comparative Study on TIBA Imputation Methods in FCMdd-Based Linear Clustering with Relational Data
title_sort comparative study on tiba imputation methods in fcmdd-based linear clustering with relational data
publisher Hindawi Limited
series Advances in Fuzzy Systems
issn 1687-7101
1687-711X
publishDate 2011-01-01
description Relational fuzzy clustering has been developed for extracting intrinsic cluster structures of relational data and was extended to a linear fuzzy clustering model based on Fuzzy c-Medoids (FCMdd) concept, in which Fuzzy c-Means-(FCM-) like iterative algorithm was performed by defining linear cluster prototypes using two representative medoids for each line prototype. In this paper, the FCMdd-type linear clustering model is further modified in order to handle incomplete data including missing values, and the applicability of several imputation methods is compared. In several numerical experiments, it is demonstrated that some pre-imputation strategies contribute to properly selecting representative medoids of each cluster.
url http://dx.doi.org/10.1155/2011/265170
work_keys_str_mv AT takeshiyamamoto acomparativestudyontibaimputationmethodsinfcmddbasedlinearclusteringwithrelationaldata
AT katsuhirohonda acomparativestudyontibaimputationmethodsinfcmddbasedlinearclusteringwithrelationaldata
AT akiranotsu acomparativestudyontibaimputationmethodsinfcmddbasedlinearclusteringwithrelationaldata
AT hidetomoichihashi acomparativestudyontibaimputationmethodsinfcmddbasedlinearclusteringwithrelationaldata
AT takeshiyamamoto comparativestudyontibaimputationmethodsinfcmddbasedlinearclusteringwithrelationaldata
AT katsuhirohonda comparativestudyontibaimputationmethodsinfcmddbasedlinearclusteringwithrelationaldata
AT akiranotsu comparativestudyontibaimputationmethodsinfcmddbasedlinearclusteringwithrelationaldata
AT hidetomoichihashi comparativestudyontibaimputationmethodsinfcmddbasedlinearclusteringwithrelationaldata
_version_ 1725490726894043136

Similar Items