Analysis of the Stability of Clustering Structures Based on the MCDM Methods

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27510%2F17%3A10236888" target="_blank" >RIV/61989100:27510/17:10236888 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Analysis of the Stability of Clustering Structures Based on the MCDM Methods
Original language description
The measurement of regional development and the clustering of regions are the problem of high importance, especially in the context of EU Regional policy. Properly defined a lagging cluster of regions is the main recipient of the Structural and Cohesion Funds. Traditional clustering methods fail to rank regions what makes these methods not applicable for the problem of recipients' determination. In the current measurement practice MCDM methods are considered to be a suitable ranking tool. Meanwhile MCDM methods are not meant to cluster the alternatives considering similarities between all alternatives within the cluster in terms of all criteria. The quasi-clustering can be done, which is based on the consideration of distances to the etalon established by MCDM methods. In this research the stability analysis has been suggested as a tool to validate the number of quasi-clusters, to identify the most stable structure and to define the most suitable MCDM method for the clustering. Two MCDM methods were chosen to measure the regional development, in particular VIKOR and Hellwig's. By the stability analysis the VIKOR method was defined as the most suitable for clustering. Clustering Stability analysis was applied for the 35 NUTS 2 regions from Visegrad group.
Czech name
—
Czech description
—

Publication year
2017
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
IBIMA 2017 : proceedings of the 30th international business information management association conference : November 8-9, 2017, Madrid, Spain
ISBN
978-0-9860419-9-0
ISSN
—
e-ISSN
neuvedeno
Number of pages
9
Pages from-to
3629-3637
Publisher name
International Business Information Management Association
Place of publication
Norristown
Event location
Madrid
Event date
Nov 8, 2017
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
000443640502112

Similar results(10)