Naslov
Influence of Distance Measure Choice On the Results of Hierarchical Clustering

Autori
Pinjušić, Sofija ; Vranić, Mihaela

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of 33rd international convention on information and communication technology, electronics and microelectronics - MIPRO 2010, Vol. V. conferences: DE, ISS, miproBIS, GLGPS, SP / Čišić, Dragan ; Hutinski, Željko ; Baranović, Mirta ; Mauher, Mladen ; Pletikosa , Marko - Zagreb : Hrvatska udruga za informacijsku i komunikacijsku tehnologiju, elektroniku i mikroelektroniku - MIPRO, 2010, 219-224

ISBN
978-953-233-055-7

Skup
33rd international convention on information and communication technology, electronics and microelectronics - MIPRO 2010 - Business Intelligence Systems

Mjesto i datum
Opatija, Hrvatska, 24.05.2010. - 28.05.2010

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Hierarchical clustering; Distance measure; Variable types

Sažetak
Hierarchical clustering method is used to assign observations into clusters which are further connected to form a hierarchical structure. Observations in the same cluster are close together according to the predetermined distance measure while observations belonging to distinct clusters are afar. This paper presents an implementation of specific distance measure used to calculate distances between observations which are described by a mixture of variable types. Data mining tool ‘Orange’ offers ways to program new modules in an efficient manner so it was used for implementation, testing, data processing and result visualization. Finally, a comparison was made between results obtained by previously available widget and output of newly programmed widget which employed new variable types and new distance measure.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika

POVEZANOST RADA