The determination of tariff classes is a relevant part of rate making process. Cluster analysistechniques have a broad range of applications in solving classification problems, but theyhave been rarely applied in insurance field because of some intrinsic difficulties. However,suitable implementations of unsupervised neural networks seem to overcome thesedifficulties. In this paper we review some families of clustering techniques and propose aneural solution to determine tariff classes. Classification results obtained with both classicaland neural clustering techniques are compared. The experiments seem to encourage furtherinvestigation on neural tools.
Determination of tariff classes: cluster analysis methods and unsupervised neural networks
GIULINI, SAVERIO;
1997-01-01
Abstract
The determination of tariff classes is a relevant part of rate making process. Cluster analysistechniques have a broad range of applications in solving classification problems, but theyhave been rarely applied in insurance field because of some intrinsic difficulties. However,suitable implementations of unsupervised neural networks seem to overcome thesedifficulties. In this paper we review some families of clustering techniques and propose aneural solution to determine tariff classes. Classification results obtained with both classicaland neural clustering techniques are compared. The experiments seem to encourage furtherinvestigation on neural tools.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.