Application of self organizing maps in understanding cultural factors and change

No Thumbnail Available
Journal Title
Journal ISSN
Volume Title
School of Electrical Engineering | Master's thesis
Checking the digitized thesis and permission for publishing
Instructions for the author
Author
Date
2012
Department
Tietoliikenne- ja tietoverkkotekniikan laitos
Major/Subject
Tietoverkkotekniikka
Mcode
S-38
Degree programme
Language
en
Pages
60 s. + liitt. 13
Series
Abstract
Human society can be viewed as a group of people who share some common or deviating culture. Societies are composed of cultures and in this work their properties and differences are studied and modelled based on the theories of Cultural Dimensioning. We will focus on dimensioning principles established by Geert Hofstede, Shalom Schwartz and World Values Survey that are commonly used as a basis for cultural dimensioning studies. Important feature of our work is to study the Self Organizing Map (SOM) in understanding the relating modelling and data representation challenges. Efficient visualization of this data requires intelligent reduction of dimensions or recognition of self-similarities within the datasets. SOM is one of the algorithms that can be used for the analysis and visualization of this high dimensional data resulting classification and clustering of the datasets. To analyse the cultural data in practice, we apply the "SOM Toolbox" running in Matlab that provides functions for building and analysing the SOMs. The Hofstede's cultural data is then used as the input and bases for SOM training. The respective outputs are then illustrated by plots reflecting the mapped cultural dimensions. In this study we also compare the Hofstede's SOM clusters to Schwartz's and World Values Survey's respective SOM clusters. As a forthcoming research we may suggest development of a graphical user interface for easier data set selection editing and provision for more flexible and insightful SOM training and analysis.
Description
Supervisor
Korhonen, Timo
Keywords
culture, self organizing map, Kohonen algorithm
Other note
Citation
Permanent link to this item
https://urn.fi/URN:NBN:fi:aalto-2020122858741
Collections
[dipl] Sähkötekniikan korkeakoulu / ELEC