Naslov
Visualization of data in lower dimensional space

Autori
Dobša, Jasminka ; Kiers, Henk A.L.

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
Book of abstracts of 23rd International Symposium on Biometrics BIOSTAT 2017 / Jazbec, Anamarija ; Pecina, Marija ; Sonicki, Zdenko ; Šimić, Diana ; Vedriš, Mislav - Zagreb, 2017, 22-22

Skup
23rd International Symposium on Biometrics BIOSTAT 2017

Mjesto i datum
Šibenik, Hrvatska, 07.06.2017. - 10.06.2017

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
reduction of dimension ; Reduced k-means method ; Factorial k-means method ; visualization of data

Sažetak
Reduction of dimensions of data representation can be benecial for diverse statistics or machine learning tasks. It can be achieved by selection of existing variables or by extraction of new variables. Our aim is to use methods for dimensionality reduction by extraction of new variables in order to represent data in a way which will enable better performance of clustering and automatic classication of objects. We will use methods of Reduced k-means, Factorial k-means and propose a modied method of Reduced k-means with penalization to project data in lower dimensional space and visualize it. Reduced k- means and Factorial k-means methods are methods that aim at reduction of variables by capturing the clustering structure present in the data. The aim of research is to test usefulness of these methods for clustering and classication of objects by visualization in lower dimensional space.

Izvorni jezik
Engleski

Znanstvena područja
Informacijske i komunikacijske znanosti

POVEZANOST RADA