Variational learning of a shifted scaled Dirichlet model with component splitting approach

Author(s):  
Narges Manouchehri ◽  
Oumayma Dalhoumi ◽  
Manar Amayri ◽  
Nizar Bouguila
2011 ◽  
Author(s):  
Malcolm Wright ◽  
Lara Stocchi ◽  
Carl Driesener

Author(s):  
JOAQUÍN ABELLÁN ◽  
ANDRÉS R. MASEGOSA

In this paper, we present the following contributions: (i) an adaptation of a precise classifier to work on imprecise classification for cost-sensitive problems; (ii) a new measure to check the performance of an imprecise classifier. The imprecise classifier is based on a method to build simple decision trees that we have modified for imprecise classification. It uses the Imprecise Dirichlet Model (IDM) to represent information, with the upper entropy as a tool for splitting. Our new measure to compare imprecise classifiers takes errors into account. Thus far, this has not been considered by other measures for classifiers of this type. This measure penalizes wrong predictions using a cost matrix of the errors, given by an expert; and it quantifies the success of an imprecise classifier based on the cardinal number of the set of non-dominated states returned. To compare the performance of our imprecise classification method and the new measure, we have used a second imprecise classifier known as Naive Credal Classifier (NCC) which is a variation of the classic Naive Bayes using the IDM; and a known measure for imprecise classification.


2021 ◽  
Author(s):  
Ahmed Rebei ◽  
Oumayma Dalhoumi ◽  
Narges Manouchehri ◽  
Ali Baghdadi ◽  
Manar Amayri ◽  
...  

Author(s):  
Shamim Chowdhury ◽  
Alicia Barker ◽  
Giang Trinh ◽  
Larry Lockshin
Keyword(s):  

2019 ◽  
Vol 224 ◽  
pp. 01004
Author(s):  
Oleg Yaremko ◽  
Natalia Yaremko

In this paper, potential fields in areas with plane and circular symmetry have been studied. In this case, the field potential is defined as the sum of solutions of Dirichlet model boundary value problems. The reflection method is used for the modeling of stationary thermal fields in multilayer media. By applying the reflection method, we found analytical solutions of boundary value problems with boundary conditions of the fourth kind for the Laplace equations and developed new computational algorithms. The developed algorithms can be easily implemented and transformed into a computer code. First of all, these algorithms implement consistent solutions of the Dirichlet problems for the model domains that allows using libraries of subroutines. Secondly, they have high algorithmic efficiency. It has been shown that reflection method is identical to the method of transformation operators and proved that transformation operator can be decomposed into a series of successive reflections from the external and internal boundaries. Finally, a physical interpretation of the reflection method has been discussed in detail.


Sign in / Sign up

Export Citation Format

Share Document