A Proposal for a Discriminant Analysis Based on the Results of a Preliminary Fuzzy Clustering
The common classification techniques are designed for a rigid (even if probabilistic) allocation of each unit into one of several groups. Nevertheless the dissimilarity among combined units often leads to consider the opportunity of assigning each of them to more than a single group with different degrees of membership. The same logic can be applied in attributing a new observation to previously identified fuzzy groups. This paper precisely presents a proposal for a discriminant analysis, structured by regressing the degrees of membership to every groups of each unit on the same variables used in a preliminary clustering. Such a proposal, initially conceived to assign new customers to defined groups for Customer Relationship Management (CRM) purposes, is now tested in an applicative case concerning the entrepreneurial propensity of the sampled provinces of Central and Southern Italy, in which an iterative fuzzy k-means method is preliminary used to split them into an optimal number of homogeneous groups.
KeywordsFuzzy discriminant analysis fuzzy clustering linear regression model customer relationship management entrepreneurial propensity
Unable to display preview. Download preview PDF.
- 1.Campobasso, F., Montrone, S., Perchinunno, P., Fanizzi, A.: A Fuzzy Approach to the Small Area Estimation of Poverty in Italy. In: Phillips-wren, G., Nakamatsu, K., Jain, L.C., Howlett, R.J. (eds.) Advances in Intelligent Decision Technologies. SIST, vol. 4, pp. 309–318. Springer, Heidelberg (2010)CrossRefGoogle Scholar
- 3.UnionCamera, Italain atlas of the competitiveness of provinces and regions, http://www.unioncamere.gov.it/Atlante/
- 4.National Institute for Foreign Trade, Italy in the International Economy - Summary Report for 2008-2009, http://www.slideshare.net/Maryss82/situazione-di-internazionalizzazione-italia-2009-ice