Pregled bibliografske jedinice broj: 843420
SOM Ward clustering approach for client segmentation in leasing industry
SOM Ward clustering approach for client segmentation in leasing industry // 21st Young Statisticians Meeting - YSM ; Programme, Abstracts, Participants / Batagelj, Vladimir ; Ferligoj, Anuška (ur.).
Ljubljana: CMI, FDV, University of Ljubljana, 2016. str. 22-22 (pozvano predavanje, nije recenziran, sažetak, ostalo)
CROSBI ID: 843420 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
SOM Ward clustering approach for client segmentation in leasing industry
Autori
Pivar, Jasmina
Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, ostalo
Izvornik
21st Young Statisticians Meeting - YSM ; Programme, Abstracts, Participants
/ Batagelj, Vladimir ; Ferligoj, Anuška - Ljubljana : CMI, FDV, University of Ljubljana, 2016, 22-22
Skup
21st Young Statisticians Meeting
Mjesto i datum
Piran, Slovenija, 04.11.2016. - 06.11.2016
Vrsta sudjelovanja
Pozvano predavanje
Vrsta recenzije
Nije recenziran
Ključne riječi
leasing ; market segmentation ; clustering ; self-orginizing maps ; SOM ; SOM-Ward
Sažetak
Leasing companies seek for better ways of reaching the clients and improving the effectiveness of their campaigns. One way to do this is to target potential clients with the particular attributes. Market segmentation can be done by using different approaches based on defined several criteria. For example, leasing company may divide the larger market into segments by using demographic and operational criteria. Established criteria may include client's size and address, leasing object, the amount of rent, reference interest rate and so on. Continuously increasing amounts of data in databases are providing companies with the opportunity to gain insight into customer behaviour and predict future agreement status, for example, whether the agreement will be expiry regularly, or there is a possibility for a fraud. The purpose of this study is to determine clients segments, common client profiles and identify segments in which frauds are most possible. SOM-Ward clustering done by using Viscovery SOMine is a useful tool for cluster analysis. We are going to adopt SOM- Ward clustering method to segment the customer base of a Croatian leasing company. The used database contains data on all leasing agreements that were active or completed at the time of running the report, including more than 40 thousands records of raw data. This study proposes to segment the client base according to the demographic and behavioural characteristics of the clients, and operational characteristics related to the leasing agreement.
Izvorni jezik
Engleski
Znanstvena područja
Ekonomija, Informacijske i komunikacijske znanosti
