Bacelar-Nicolau, H. (1985): The aOEnity coeOEcient in Cluster Analysis. Methods of Operations Research, vol.53, 507-512.  Home/Search   Document Not in Database   Summary   Related Articles

....b , 1993] and developped by him and his collaborators. It is implemented in the computer program CHAVL [Lerman, Peter Leredde, 1993 1994] We have used a speci c INRIA Reducing the number of binary splits, in Decision Tree 17 form for the agregation criterion, called AV L 0:5 , and proposed by [Bacelar Nicolau, 1985]. The relevancy of this methodology is motivated by the following reasons ; i) It is underlined by a very general constructive method of probabilistic similarity indices between combinatorial or logical structures, on a given object set O. These structures are interpreted in terms of relations on ....

.... and m = card(H) Lerman 1970,1981] The latter index corresponds to what we call the pure form of the ivraisemblance du lien maximalj (imaximal link likelihoodj) criterion, that we can denote by V L 1 (G; H) p(G; H) l Thetam Following the works of Nicolau and Bacelar Nicolau ( Nicolau,1981] [Bacelar Nicolau,1985]) we have suggested the family of criteria : V L (G; H) p(G; H) l Thetam) where is a real parameter between 0 and 1. This family goes from the maximal link ( 0) to the pure form of the maximam link likelihood criterion ( 1) For reasons of accuracy in computing, we consider ....

Bacelar-Nicolau, H. (1985): The aOEnity coeOEcient in Cluster Analysis. Methods of Operations Research, vol.53, 507-512.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC