engleski

Application of Kohonen Self-Organizing Maps (SOM) Based Clustering for the Assessment of Religious Motivation

We investigated possible application of the Kohonen self-organizing maps (SOM) to the social sciences data clustering, and compare the results of the procedure to the principal components analysis (PCA) and hierarchical clustering methods. Real dataset consisting of the responses of 473 individuals to the questions of the Hoge intrinsic religious motivation scale was used. It is concluded that combined approach to the exploratory data analysis by means of SOM, PCA and hierarchical clustering gives a more clear picture of variable interrelations and grouping. Consequently it may be of help for the exploratory data analyses in social (and natural) sciences.

self; organizing; maps; principal; component; analysis; data; explaratory

nije evidentirano