Naslov
Segmentation in banking using self-organizing maps: A case study of business customers

Autori
Juković, Sandro ; Pejić Bach, Mirjana ; Dumičić, Ksenija ; Šarlija, Nataša

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

Izvornik
Conference Proceedings of 6th International Conference of the School of Economics and Business Beyond the Economic Crisis: Lessons Learned and Challenges Ahead / Mehić, Eldin - Sarajevo : Ekonomski fakultet u Sarajevu, 2012, 767-778

Skup
6th International Conference of the School of Economics and Business Beyond the Economic Crisis: Lessons Learned and Challenges Ahead

Mjesto i datum
Sarajevo, Bosna i Hercegovina, 12.10.2012. - 13.10.2012

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Self-organizing maps; segmentation; banking; neural networks

Sažetak
The aim of this paper is to present the theoretical basis of SOM and to show its application in the field of market segmentation. The paper explains the SOM-Ward algorithm which is implemented in software Viscovery SOMine. Data set for this research is consisted of banks’ business customers of several banks in Croatia. The cluster analysis using the SOMine software is applied. Segmentation of business customers of banks in Croatia is presented after determining the segmentation criteria based on behavioral, demographic, personal, operational, situational, and criteria related to the purchase of certain services. The results of the analysis are presented and interpreted as three segments. Aforementioned segments differ with respect to the attributes of the foreign trade operations (import/export), annual income, origin of capital, the views on the loan selection, hiring plans, etc. Analyzed segments can be used by banks for deciding whether to take further marketing activities and in which direction.

Izvorni jezik
Engleski

Znanstvena područja
Informacijske i komunikacijske znanosti

POVEZANOST RADA