Afif Masmoudi - Profile on Academia.edu (original) (raw)

Papers by Afif Masmoudi

Journal of Computational and Applied Mathematics, 2015

Currently, Bayesian Networks (BNs) have become one of the most complete, self-sustained and coher... more Currently, Bayesian Networks (BNs) have become one of the most complete, self-sustained and coherent formalisms used for knowledge acquisition, representation and application through computer systems. However, learning of BNs structures from data has been shown to be an NP-hard problem. It has turned out to be one of the most exciting challenges in machine learning. In this context, the present work's major objective lies in setting up a further solution conceived to be a remedy for the intricate algorithmic complexity imposed during the learning of BN-structure with a massively-huge data backlog.

Journal of Mathematical Modelling and Algorithms in Operations Research, 2014

The use of Bayesian Networks (BNs) as classifiers in different application fields has recently wi... more The use of Bayesian Networks (BNs) as classifiers in different application fields has recently witnessed a noticeable growth. Yet, using the Naïve Bayes application, and even the augmented Naïve Bayes, to classifier-structure learning, has been vulnerable to some extent, which accounts for the resort of experts to other more sophisticated types of algorithms. Consequently, the use of such algorithms has paved the way for raising the problem of super-exponential increase in computational complexity of the Bayesian classifier learning structure, with the increasing number of descriptive variables. In this context, the main objective of our present work lies in trying to conceive further solutions to solve the problem of the intricate algorithmic complexity imposed during the learning of Bayesian classifiers structure through the use of sophisticated algorithms. Our results revealed that the newly suggested approach allows us to considerably reduce the execution time of the Bayesian classifier structure learning without any information loss.

International Journal of Computational Intelligence Systems, 2014

The Bayesian estimation of the conditional Gaussian parameter needs to define several a priori pa... more The Bayesian estimation of the conditional Gaussian parameter needs to define several a priori parameters. The proposed approach is free from this definition of priors. We use the Implicit estimation method for learning from observations without a prior knowledge. We illustrate the interest of such an estimation method by giving first the Bayesian Expectation A Posteriori estimator for conditional Gaussian parameters. Then, we describe the Implicit estimators for the same parameters. Moreover, an experimental study is proposed in order to compare both approaches.

Pakistan Journal of Statistics and Operation Research, Dec 5, 2023

Neurocomputing, Jul 1, 2019

Iet Image Processing, Dec 1, 2014

Journal of Electronic Imaging, Apr 1, 2010

Applied sciences, Feb 24, 2023

In this paper, we consider a coded cooperative communication network with multiple energy-harvest... more In this paper, we consider a coded cooperative communication network with multiple energy-harvesting (EH) relays. In order to adequately address the problem of error propagation due to the erroneous decoding at the relays, as in the case of conventional decode and forward (DF) relaying protocol, we propose coded cooperative schemes with hard information relaying (HIR) and soft information relaying (SIR) strategies. The performance of the relayed communication with EH relay depends crucially on the channel decoding capability at the relay, channel gains at the source-relay and relay-destination links, and ultimately on the power-splitting ratio of the relay EH receiver. The exact closed-form expression for the outage probability performance of the coded cooperative scheme with HIR strategy and relay selection (CC-HIR-RS) is derived for both cases, namely for constant and optimal power-splitting ratios. Concerning the coded cooperative scheme with SIR strategy, a Rayleigh Gaussian log likelihood ratio-based model is used to describe the soft estimated symbols at the output of the relay soft encoder. Directives are provided to determine the model parameters, and, accordingly, the signal-to-noise ratio (SNR) of the equivalent one-hop relaying channel is derived. A closed-form expression for the outage probability performance of the proposed coded cooperative scheme with SIR and relay selection (CC-SIR-RS) is derived. In addition, a fuzzy logic-based power-splitting scheme in EH relay applying SIR is proposed. The fading coefficients of the source-relay and relay-destination links and distance between source and relay node are considered as input parameters of the fuzzy logic system to obtain an appropriate power-splitting ratio that leads to a quasi-optimal SNR of the equivalent end-to-end channel. Monte Carlo simulations are presented to demonstrate the validity of the analytical results, and a comparison between the performance of the CC-HIR-RS scheme with constant and optimized power-splitting ratios and that of the CC-SIR-RS scheme with constant and fuzzy logic-based power-splitting ratios is provided.

Entropy, 2022

The central focus of this paper is upon the alleviation of the boundary problem when the probabil... more The central focus of this paper is upon the alleviation of the boundary problem when the probability density function has a bounded support. Mixtures of beta densities have led to different methods of density estimation for data assumed to have compact support. Among these methods, we mention Bernstein polynomials which leads to an improvement of edge properties for the density function estimator. In this paper, we set forward a shrinkage method using the Bernstein polynomial and a finite Gaussian mixture model to construct a semi-parametric density estimator, which improves the approximation at the edges. Some asymptotic properties of the proposed approach are investigated, such as its probability convergence and its asymptotic normality. In order to evaluate the performance of the proposed estimator, a simulation study and some real data sets were carried out.

Arabian Journal of Geosciences, 2020

AStA Advances in Statistical Analysis, 2020

Filomat, 2018

This paper is concerned with a class of exponential dispersion distributions. We particularly foc... more This paper is concerned with a class of exponential dispersion distributions. We particularly focused on the mixture models, which represent an extension of the Gaussian distribution. It should be noted that the parameters estimation of mixture distributions is an important task in statistical processing. In order to estimate the parameters vector, we proposed a formulation of the Expectation-Maximization algorithm (EM) under exponential dispersion mixture distributions. Also, we developed a hybrid algorithm called Expectation-Maximization and Method of moments algorithm (EMM). Under mild regularity, several convergence results of the EMM algorithm were obtained. Through simulation studies, the robustness of the EMM was proved and the strong consistency of the EMM sequence appeared when the data set size and the number of iterations tend to infinity.

2017 25th European Signal Processing Conference (EUSIPCO), 2017

Asian Economic and Financial Review, 2017

This paper deals with the issue of predicting customers' default payment. The Bayesian network cr... more This paper deals with the issue of predicting customers' default payment. The Bayesian network credit model is applied for the prediction and classification of personal loans with regard to credit worthiness. Referring to credit experts and using K2 algorithm for learning structure, we set up the dependency conditional relations between variables that explain default payments. Then, the parametric learning is adopted to detect conditional probabilities of customers' default payment. The parameters are estimated on the basis of real personal loan data obtained from a Tunisian commercial bank. The Bayesian network analysis has revealed that customers' age, gender, type of credit, professional status, and monthly repayment burden and credit duration have an important predictive power for the detection of customers' default payment. Therefore, our findings allow providing an effective decision support system for banks in order to detect and reduce the rate of bad borrowers through the use of a Bayesian Network model.

Communications in Statistics - Simulation and Computation, 2017

AEU - International Journal of Electronics and Communications, 2016

Communications in Computer and Information Science, 2012

It is a well-known fact that the Bayesian Networks' (BNs) use as classifiers in different fields ... more It is a well-known fact that the Bayesian Networks' (BNs) use as classifiers in different fields of application has recently witnessed a noticeable growth. Yet, the Naïve Bayes' application, and even the augmented Naïve Bayes', to classifier-structure learning, has been vulnerable to certain limits, which explains the practitioners' resort to other more sophisticated types of algorithms. Consequently, the use of such algorithms has paved the way for raising the problem of super-exponential increase in computational complexity of the Bayesian classifier learning structure, with the increasing number of descriptive variables. In this context, the present work's major objective lies in setting up a further solution whereby a remedy can be conceived for the intricate algorithmic complexity imposed during the learning of Bayesian classifiers' structure with the use of sophisticated algorithms. Noteworthy, the present paper's framework is organized as follows. We start, in the first place, by to propose a novel approach designed to reduce the algorithmic complexity without engendering any loss of information when learning the structure of a Bayesian classifier. We, then, go on to test our approach on a car diagnosis and a Lymphography diagnosis databases. Ultimately, an exposition of our conducted work's interests will be a closing step to this work.

Acta Scientiarum. Agronomy, 2012

Multimedia Tools and Applications, 2014

2013 IEEE 17th International Conference on Intelligent Engineering Systems (INES), 2013

Journal of Computational and Applied Mathematics, 2015

Currently, Bayesian Networks (BNs) have become one of the most complete, self-sustained and coher... more Currently, Bayesian Networks (BNs) have become one of the most complete, self-sustained and coherent formalisms used for knowledge acquisition, representation and application through computer systems. However, learning of BNs structures from data has been shown to be an NP-hard problem. It has turned out to be one of the most exciting challenges in machine learning. In this context, the present work's major objective lies in setting up a further solution conceived to be a remedy for the intricate algorithmic complexity imposed during the learning of BN-structure with a massively-huge data backlog.

Journal of Mathematical Modelling and Algorithms in Operations Research, 2014

The use of Bayesian Networks (BNs) as classifiers in different application fields has recently wi... more The use of Bayesian Networks (BNs) as classifiers in different application fields has recently witnessed a noticeable growth. Yet, using the Naïve Bayes application, and even the augmented Naïve Bayes, to classifier-structure learning, has been vulnerable to some extent, which accounts for the resort of experts to other more sophisticated types of algorithms. Consequently, the use of such algorithms has paved the way for raising the problem of super-exponential increase in computational complexity of the Bayesian classifier learning structure, with the increasing number of descriptive variables. In this context, the main objective of our present work lies in trying to conceive further solutions to solve the problem of the intricate algorithmic complexity imposed during the learning of Bayesian classifiers structure through the use of sophisticated algorithms. Our results revealed that the newly suggested approach allows us to considerably reduce the execution time of the Bayesian classifier structure learning without any information loss.

International Journal of Computational Intelligence Systems, 2014

The Bayesian estimation of the conditional Gaussian parameter needs to define several a priori pa... more The Bayesian estimation of the conditional Gaussian parameter needs to define several a priori parameters. The proposed approach is free from this definition of priors. We use the Implicit estimation method for learning from observations without a prior knowledge. We illustrate the interest of such an estimation method by giving first the Bayesian Expectation A Posteriori estimator for conditional Gaussian parameters. Then, we describe the Implicit estimators for the same parameters. Moreover, an experimental study is proposed in order to compare both approaches.

Pakistan Journal of Statistics and Operation Research, Dec 5, 2023

Neurocomputing, Jul 1, 2019

Iet Image Processing, Dec 1, 2014

Journal of Electronic Imaging, Apr 1, 2010

Applied sciences, Feb 24, 2023

In this paper, we consider a coded cooperative communication network with multiple energy-harvest... more In this paper, we consider a coded cooperative communication network with multiple energy-harvesting (EH) relays. In order to adequately address the problem of error propagation due to the erroneous decoding at the relays, as in the case of conventional decode and forward (DF) relaying protocol, we propose coded cooperative schemes with hard information relaying (HIR) and soft information relaying (SIR) strategies. The performance of the relayed communication with EH relay depends crucially on the channel decoding capability at the relay, channel gains at the source-relay and relay-destination links, and ultimately on the power-splitting ratio of the relay EH receiver. The exact closed-form expression for the outage probability performance of the coded cooperative scheme with HIR strategy and relay selection (CC-HIR-RS) is derived for both cases, namely for constant and optimal power-splitting ratios. Concerning the coded cooperative scheme with SIR strategy, a Rayleigh Gaussian log likelihood ratio-based model is used to describe the soft estimated symbols at the output of the relay soft encoder. Directives are provided to determine the model parameters, and, accordingly, the signal-to-noise ratio (SNR) of the equivalent one-hop relaying channel is derived. A closed-form expression for the outage probability performance of the proposed coded cooperative scheme with SIR and relay selection (CC-SIR-RS) is derived. In addition, a fuzzy logic-based power-splitting scheme in EH relay applying SIR is proposed. The fading coefficients of the source-relay and relay-destination links and distance between source and relay node are considered as input parameters of the fuzzy logic system to obtain an appropriate power-splitting ratio that leads to a quasi-optimal SNR of the equivalent end-to-end channel. Monte Carlo simulations are presented to demonstrate the validity of the analytical results, and a comparison between the performance of the CC-HIR-RS scheme with constant and optimized power-splitting ratios and that of the CC-SIR-RS scheme with constant and fuzzy logic-based power-splitting ratios is provided.

Entropy, 2022

The central focus of this paper is upon the alleviation of the boundary problem when the probabil... more The central focus of this paper is upon the alleviation of the boundary problem when the probability density function has a bounded support. Mixtures of beta densities have led to different methods of density estimation for data assumed to have compact support. Among these methods, we mention Bernstein polynomials which leads to an improvement of edge properties for the density function estimator. In this paper, we set forward a shrinkage method using the Bernstein polynomial and a finite Gaussian mixture model to construct a semi-parametric density estimator, which improves the approximation at the edges. Some asymptotic properties of the proposed approach are investigated, such as its probability convergence and its asymptotic normality. In order to evaluate the performance of the proposed estimator, a simulation study and some real data sets were carried out.

Arabian Journal of Geosciences, 2020

AStA Advances in Statistical Analysis, 2020

Filomat, 2018

This paper is concerned with a class of exponential dispersion distributions. We particularly foc... more This paper is concerned with a class of exponential dispersion distributions. We particularly focused on the mixture models, which represent an extension of the Gaussian distribution. It should be noted that the parameters estimation of mixture distributions is an important task in statistical processing. In order to estimate the parameters vector, we proposed a formulation of the Expectation-Maximization algorithm (EM) under exponential dispersion mixture distributions. Also, we developed a hybrid algorithm called Expectation-Maximization and Method of moments algorithm (EMM). Under mild regularity, several convergence results of the EMM algorithm were obtained. Through simulation studies, the robustness of the EMM was proved and the strong consistency of the EMM sequence appeared when the data set size and the number of iterations tend to infinity.

2017 25th European Signal Processing Conference (EUSIPCO), 2017

Asian Economic and Financial Review, 2017

This paper deals with the issue of predicting customers' default payment. The Bayesian network cr... more This paper deals with the issue of predicting customers' default payment. The Bayesian network credit model is applied for the prediction and classification of personal loans with regard to credit worthiness. Referring to credit experts and using K2 algorithm for learning structure, we set up the dependency conditional relations between variables that explain default payments. Then, the parametric learning is adopted to detect conditional probabilities of customers' default payment. The parameters are estimated on the basis of real personal loan data obtained from a Tunisian commercial bank. The Bayesian network analysis has revealed that customers' age, gender, type of credit, professional status, and monthly repayment burden and credit duration have an important predictive power for the detection of customers' default payment. Therefore, our findings allow providing an effective decision support system for banks in order to detect and reduce the rate of bad borrowers through the use of a Bayesian Network model.

Communications in Statistics - Simulation and Computation, 2017

AEU - International Journal of Electronics and Communications, 2016

Communications in Computer and Information Science, 2012

It is a well-known fact that the Bayesian Networks' (BNs) use as classifiers in different fields ... more It is a well-known fact that the Bayesian Networks' (BNs) use as classifiers in different fields of application has recently witnessed a noticeable growth. Yet, the Naïve Bayes' application, and even the augmented Naïve Bayes', to classifier-structure learning, has been vulnerable to certain limits, which explains the practitioners' resort to other more sophisticated types of algorithms. Consequently, the use of such algorithms has paved the way for raising the problem of super-exponential increase in computational complexity of the Bayesian classifier learning structure, with the increasing number of descriptive variables. In this context, the present work's major objective lies in setting up a further solution whereby a remedy can be conceived for the intricate algorithmic complexity imposed during the learning of Bayesian classifiers' structure with the use of sophisticated algorithms. Noteworthy, the present paper's framework is organized as follows. We start, in the first place, by to propose a novel approach designed to reduce the algorithmic complexity without engendering any loss of information when learning the structure of a Bayesian classifier. We, then, go on to test our approach on a car diagnosis and a Lymphography diagnosis databases. Ultimately, an exposition of our conducted work's interests will be a closing step to this work.

Acta Scientiarum. Agronomy, 2012

Multimedia Tools and Applications, 2014

2013 IEEE 17th International Conference on Intelligent Engineering Systems (INES), 2013