Drift Detection and Model Selection Algorithms: Concept and Experimental Evaluation

2012 ◽  
pp. 558-568
Author(s):  
Piotr Cal ◽  
Michał Woźniak
Keyword(s):  
Model Selection ◽  
Experimental Evaluation ◽  
Selection Algorithms

ENTROPY PENALIZED AUTOMATED MODEL SELECTION ON GAUSSIAN MIXTURE

2004 ◽  
Vol 18 (08) ◽  
pp. 1501-1512 ◽  
Author(s):  
JINWEN MA ◽  
TAIJUN WANG
Keyword(s):  
Model Selection ◽  
Likelihood Estimation ◽  
Gaussian Mixture ◽  
Mixture Modeling ◽  
Selection Algorithm ◽  
Penalized Maximum Likelihood ◽  
Gaussian Mixture Modeling ◽  
Powerful Approach ◽  
Selection Algorithms

Gaussian mixture modeling is a powerful approach for data analysis and the determination of the number of Gaussians, or clusters, is actually the problem of Gaussian mixture model selection which has been investigated from several respects. This paper proposes a new kind of automated model selection algorithm for Gaussian mixture modeling via an entropy penalized maximum-likelihood estimation. It is demonstrated by the experiments that the proposed algorithm can make model selection automatically during the parameter estimation, with the mixing proportions of the extra Gaussians attenuating to zero. As compared with the BYY automated model selection algorithms, it converges more stably and accurately as the number of samples becomes large.

scholarly journals USING MODEL SELECTION ALGORITHMS TO OBTAIN RELIABLE COEFFICIENT ESTIMATES

2011 ◽  
Vol 27 (2) ◽  
pp. 269-296 ◽  
Author(s):  
Jennifer L. Castle ◽  
Xiaochuan Qin ◽  
W. Robert Reed
Keyword(s):  
Model Selection ◽  
Coefficient Estimates ◽  
Selection Algorithms

scholarly journals Model Selection when multiple imputation is used to protect confidentiality in public use data

2011 ◽  
Vol 2 (2) ◽  
Author(s):  
Satkartar K. Kinney ◽  
Jerome P. Reiter ◽  
James O. Berger
Keyword(s):  
Model Selection ◽  
Multiple Imputation ◽  
Bayes Factor ◽  
Data Augmentation ◽  
Stochastic Search Variable Selection ◽  
Multiple Imputations ◽  
Multiple Datasets ◽  
Statistical Agencies ◽  
Synthetic Datasets ◽  
Selection Algorithms

Several statistical agencies use, or are considering the use of, multiple imputation to limit the risk of disclosing respondents' identities or sensitive attributes in public use files. For example, agencies can release partially synthetic datasets, comprising the units originally surveyed with some values, such as sensitive values at high risk of disclosure, or values of key identifiers, replaced with multiple imputations. We describe how secondary analysts of such multiply-imputed datasets can implement Bayesian model selection procedures that appropriately condition on the multiple datasets and the information released by the agency about the imputation models. We illustrate by deriving Bayes factor approximations and a data augmentation step for stochastic search variable selection algorithms.

FCM-Based Model Selection Algorithms for Determining the Number of Clusters

2004 ◽  
Vol 37 (10) ◽  
pp. 2027-2037 ◽  
Author(s):  
Haojun Sun ◽  
Shengrui Wang ◽  
Qingshan Jiang
Keyword(s):  
Model Selection ◽  
Number Of Clusters ◽  
Selection Algorithms
Sign in / Sign up

Export Citation Format

Share Document