Você está aqui: Página Inicial / Seminários / 14/09/2018 - Kernel Paired Copula Bayesian Classifier - Palestrante: Anderson Ara (UFBA)

14/09/2018 - Kernel Paired Copula Bayesian Classifier - Palestrante: Anderson Ara (UFBA)

Quando 14/09/2018
das 14h00 até 16h00
Adicionar evento ao calendário vCal
iCal

Seminário Conjunto UFSCar/ICMC

Data e horário: 14/09/2018 às 14 horas

Local: Auditório Luiz Fávaro (ICMC/USP)

Título: Kernel Paired Copula Bayesian Classifier

Palestrante: Prof. Anderson Ara (UFBA)

Resumo: Bayesian networks, also known as causal networks, belief  systems, or probabilistic dependency plots emerged in the 1980s and were  applied in a wide variety of real-world activities. However, the most  common available structure estimation algorithms underlying Bayesian network classifiers are often confined to discrete or Gaussian models. Recently, Copula methodology is considered to handle classification with Bayesian networks such as Copula Network Classifiers (CNC). This article compares the CNC with a Bayesian network of copula pairs (PCCBN) considering bivariate classification problems. The predictive performance of the both methods are compared in the modeling of real and artificial datasets. Furthermore, we propose a new Bayesian classifier based on non-parametric paired copula construction (KPCCBN).

registrado em: