A new hybrid discriminative/generative model using the full-covariance multivariate generalized Gaussian mixture models

Najar, Fatma; Bourouis, Sami; Bouguila, Nizar; Belghith, Safya

doi:10.1007/s00500-019-04567-2

Cited by 14 publications

(20 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Actually, many classifiers such as SVM show great potential compared to generative models for several applications [ 30 , 31 ]. However, in most applications, the conventional SVMs kernels (i.e., linear, polynomial, RBF) [ 30 ] are not able to consider the nature of the data and it was noted that choosing conventional SVM kernels was not the right choice [ 14 ]. This disadvantage limits their performance.…”

Section: Discriminative Learning Approach Based On Ssdmmmentioning

confidence: 99%

“…It is noted that while the data follows a non-Gaussian distribution in nature, the Gaussian model can give a weak performance. Noticing this fact, various mixture models have been proposed in the literature and some distinguished ones are based on the generalized Gaussian [ 14 ], Dirichlet and generalized Dirichlet [ 15 ], Beta-Liouville [ 15 ], t-student distributions, etc. For instance, the Dirichlet mixture and its extensions (like generalized Dirichlet) have been successfully employed and can often outperform the Gaussian model for data clustering, categorization and action recognition [ 9 , 16 , 17 , 18 , 18 ].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Discriminative Learning Approach Based on Flexible Mixture Model for Medical Data Categorization and Recognition

Alharithi

Almulihi

Bourouis

et al. 2021

Sensors

Self Cite

View full text Add to dashboard Cite

In this paper, we propose a novel hybrid discriminative learning approach based on shifted-scaled Dirichlet mixture model (SSDMM) and Support Vector Machines (SVMs) to address some challenging problems of medical data categorization and recognition. The main goal is to capture accurately the intrinsic nature of biomedical images by considering the desirable properties of both generative and discriminative models. To achieve this objective, we propose to derive new data-based SVM kernels generated from the developed mixture model SSDMM. The proposed approach includes the following steps: the extraction of robust local descriptors, the learning of the developed mixture model via the expectation–maximization (EM) algorithm, and finally the building of three SVM kernels for data categorization and classification. The potential of the implemented framework is illustrated through two challenging problems that concern the categorization of retinal images into normal or diabetic cases and the recognition of lung diseases in chest X-rays (CXR) images. The obtained results demonstrate the merits of our hybrid approach as compared to other methods.

show abstract

Section: Discriminative Learning Approach Based On Ssdmmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Discriminative Learning Approach Based on Flexible Mixture Model for Medical Data Categorization and Recognition

Alharithi

Almulihi

Bourouis

et al. 2021

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…It may be important to consider the nature of the observed data when determining a particular probability distribution. For several decades Gaussian distribution has been applied as the basic distribution to solve many data analysis problems [8], however, recently several works have proved that it is not the best approximation especially when treating non-Gaussian data [9] (i.e. with restrictions of non-negativity and unit-sum).…”

Section: Introductionmentioning

confidence: 99%

“…To take advantage of both approaches, in this paper, we consider a hybrid generativediscriminative learning approach. Hybrid framework can be seen as the process of integration of prior knowledge related to the current classification into the training operation to get the highest accuracy [8], [22]- [24]. This paper focuses on increasing the classification performance by introducing a new hybrid learning approach that takes into account the benefits of both discriminative SVM kernels and generative IBLMM mixture.…”

Section: Introductionmentioning

confidence: 99%

Markov Chain Monte Carlo-Based Bayesian Inference for Learning Finite and Infinite Inverted Beta-Liouville Mixture Models

et al. 2021

Self Cite

View full text Add to dashboard Cite

Recently Inverted Beta-Liouville mixture models have emerged as an efficient paradigm for proportional positive vectors modeling and unsupervised learning. However, little attention has been devoted to investigate these generative models within discriminative classifiers. Our aim here is to reveal the structure of non-Gaussian data by generating new probabilistic SVM kernels from inverted-Beta Liouville mixture models. The inverted Beta-Liouville has a more general covariance structure and a smaller number of parameters than the inverted Dirichlet and generalized inverted Dirichlet, respectively, which makes it more practical and useful. A principled Bayesian learning algorithm is developed to accurately estimate the model's parameters. To cope with the problem of selecting the optimal number of components, we further propose a nonparametric Bayesian learning algorithm based on an extended infinite mixture model which may have better modelling and clustering capabilities than the finite model for some applications. Finally, the resulting generative model is exploited to build several efficient probabilistic SVM kernels in order to enhance the expected clustering and modeling performance. Through a number of experimental evaluations involving visual scenes classification, text categorization and texture images discrimination, we prove the merits of the proposed work.

show abstract

“…A difficult aspect when considering finite mixture models is the determining of the exact component numbers, which help to avoid the issues of over-and under-fitting and minimizing the approximation errors, especially when we are trying to model complex real-world data problems (such as multimodal data) [9]. For instance, Laplace and Normal densities fail to fit many complex shapes with multi-dimensional data and when describing the heavier tails caused by specific patterns [10,11]. In this context, non-Gaussian data modelling plays an essential role in accurate data clustering and classification.…”

Section: Introduction and Related Workmentioning

confidence: 99%

Nonparametric Bayesian Learning of Infinite Multivariate Generalized Normal Mixture Models and Its Applications

et al. 2021

Self Cite

View full text Add to dashboard Cite

This paper addresses the problem of data vectors modeling, classification and recognition using infinite mixture models, which have been shown to be an effective alternative to finite mixtures in terms of selecting the optimal number of clusters. In this work, we propose a novel approach for localized features modelling using an infinite mixture model based on multivariate generalized Normal distributions (inMGNM). The statistical mixture is learned via a nonparametric MCMC-based Bayesian approach in order to avoid the crucial problem of model over-fitting and to allow uncertainty in the number of mixture components. Robust descriptors are derived from encoding features with the Fisher vector method, which considers higher order statistics. These descriptors are combined with a linear support vector machine classifier in order to achieve higher accuracy. The efficiency and merits of the proposed nonparametric Bayesian learning approach, while comparing it to other different methods, are demonstrated via two challenging applications, namely texture classification and human activity categorization.

show abstract

A new hybrid discriminative/generative model using the full-covariance multivariate generalized Gaussian mixture models

Cited by 14 publications

References 36 publications

Discriminative Learning Approach Based on Flexible Mixture Model for Medical Data Categorization and Recognition

Discriminative Learning Approach Based on Flexible Mixture Model for Medical Data Categorization and Recognition

Markov Chain Monte Carlo-Based Bayesian Inference for Learning Finite and Infinite Inverted Beta-Liouville Mixture Models

Nonparametric Bayesian Learning of Infinite Multivariate Generalized Normal Mixture Models and Its Applications

Contact Info

Product

Resources

About