Videos by Dhafer Malouche
Description de la situation COVID-19 en Tunisie et dans le monde
7 views
Papers by Dhafer Malouche

Implications of Faithfulness in Graphical Models
Concentration graph models and covariance graph models are two of the widely studied classes of g... more Concentration graph models and covariance graph models are two of the widely studied classes of graphical models. They are specified through pairwise relationships between variables. Under suitable conditions, they can be used to read conditional independence relations at the level of sets of variables. It’s known that faithfulness property is filled when the graph allows identifying the whole set condition independence statements. This paper studies the implications of imposing the faithfulness assumption on either the covariance or concentration graphs. We demonstrate that if a probability distribution is faithful to its concentration graph. The corresponding covariance graph is a union of complete connected components, i.e., each connected component cannot have any marginal independence among its nodes. We also prove a dual result when the distribution is faithful to its covariance graph. The general implications of the results are far-reaching. First, the result formalizes the long-held notion in the graphical models’ community that faithfulness is a very restrictive assumption. Second, we show that estimation procedures in graphical models by low-order conditioning may lead to erroneous conclusions. Since these procedures effectively search for models in a very restrictive class of probability. distributions.

Eastern Mediterranean Health Journal
Background: The SF-36 is the instrument for measuring the health relatedquality of life (HRQOL) o... more Background: The SF-36 is the instrument for measuring the health relatedquality of life (HRQOL) of patients in many clinical and national studies to describe the health status of populations, by obtaining comparable data on health status internationally. Aims: This study aimed to obtain population norms for the Tunisian version of SF-36 and to assess the association between socio HRQOL scores with the demographic characteristics of the Tunisian population. Methods: Face-to-face interviews for a cross-sectional study were carried out in 2005 to collect socio demographic and environmental variables as well as self-reported quality of life. A representative sample of 6543 aged between 35 and 70 years old were selected. Results: All scores had a high level of internal consistency reliability coefficient. HRQOL score levels were associated with sociodemographic characteristics and a decrease as age increased. The averages of the physical and mental component summary were 53+/-8 for males and 47.7+/-10 for females. Conclusions: This study was the first to address the general Tunisian population. This study shed light on factors associated with HRQOL in the Tunisian context.

PLOS Neglected Tropical Diseases
Transmission of zoonotic cutaneous leishmaniasis (ZCL) depends on the presence, density and distr... more Transmission of zoonotic cutaneous leishmaniasis (ZCL) depends on the presence, density and distribution of Leishmania major rodent reservoir and the development of these rodents is known to have a significant dependence on environmental and climate factors. ZCL in Tunisia is one of the most common forms of leishmaniasis. The aim of this paper was to build a regression model of ZCL cases to identify the relationship between ZCL occurrence and possible risk factors, and to develop a predicting model for ZCL's control and prevention purposes. Monthly reported ZCL cases, environmental and bioclimatic data were collected over 6 years (2009-2015). Three rural areas in the governorate of Sidi Bouzid were selected as the study area. Cross-correlation analysis was used to identify the relevant lagged effects of possible risk factors, associated with ZCL cases. Non-parametric modeling techniques known as generalized additive model (GAM) and generalized additive mixed models (GAMM) were applied in this work. These techniques have the ability to approximate the relationship between the predictors (inputs) and the response variable (output), and express the relationship mathematically. The goodness-of-fit of the constructed model was determined by Generalized cross-validation (GCV) score and residual test. There were a total of 1019 notified ZCL cases from July 2009 to June 2015. The results showed seasonal distribution of reported ZCL cases from August to January. The model highlighted that rodent density, average temperature, cumulative rainfall and average relative humidity, with different time lags, all play role in sustaining and increasing the ZCL incidence. The GAMM model could be applied to predict the occurrence of ZCL in central Tunisia and could help for the establishment of an early warning system to control and prevent ZCL in central Tunisia.
Frontiers in Immunology, 2016

Performance measurement and quality costing in Tunisian manufacturing companies
The TQM Journal, 2016
Purpose – The purpose of this paper is to present the results of an exploratory survey on the imp... more Purpose – The purpose of this paper is to present the results of an exploratory survey on the implementation of quality costs programmes in Tunisian manufacturing companies. It aims also to observe the impact of quality cost system (QCS) adoption on operational performance by the manufacturing companies. Design/methodology/approach – An interview-based survey was employed in order to investigate the extent and consequences of implementing quality costs programmes in Tunisian manufacturing companies. A questionnaire was developed and used in the interviews across a sample of Tunisian companies. Findings – A factor analysis revealed that companies which adopted QCS have less internal and external failures costs than others. It also revealed that these companies invest more in prevention and appraisal actions than others. It showed that QCS adoption has positive consequences on the companies’ performance. Research limitations/implications – The presented results are limited by the low ...
A Property of Count Distributions in the Hinde�Dem�trio Family
Commun Statist Theor Method, 2008
Selecting test of distribution in the Hinde-Dem�trio family
Familles exponentielles associees a des fonctions pick et classification fonctions variances p(m)/(cm + d)
Http Www Theses Fr, 1997

American journal of epidemiology, 1979
All causes of death related to the two risk factors, smoking and hazardous drinking, have been re... more All causes of death related to the two risk factors, smoking and hazardous drinking, have been reviewed followed by a selection of those causes of death for which the causal role of the risk factor appears to be quasi-certain. For each cause, existing epidemiologic data were reviewed and used to determine the fraction of premature mortality which could be attributed to each factor (called the attributable fraction). This fraction was then multiplied by the corresponding Canadian premature mortality measured in terms of deaths between ages one and 70 and potential years of life lost (PYLL) between ages one and 70, which gives a higher weight to younger deaths. Of the 73,440 deaths between ages one and 70 in Canada in 1974, 12% (or 8718 deaths) were found to be attributable to current smoking and 6% (4716) to hazardous drinking. In terms of PYLL between ages one and 70, hazardous drinking ranks ahead of current smoking with 10% (or 132,044 PYLL) of the total PYLL, whereas current smok...
Méthodologie de l’analyse de la relation température–mortalité en Tunisie
Revue d'Épidémiologie et de Santé Publique, 2014
Resume : Pour detecter l'osteoporose `a partir de sa probabilite de presence, on s'inte... more Resume : Pour detecter l'osteoporose `a partir de sa probabilite de presence, on s'interesse `a estimer cette derni`ere `a partir d'un jeu de donnees o`u certaines observations peuventetre manquantes au mo-ment du calcul de la probabilite. Nous avons ici recours aux outils de ...
Modèle Graphique Ou Champ Markovien Gaussien En Image: Estimation Du Graphe De Voisinage
Le recours aux champs Markoviens en modelisation d'image a connu un grand succ`es vu qu'... more Le recours aux champs Markoviens en modelisation d'image a connu un grand succ`es vu qu'ils permettent de traduire les intercations spatiales entre les niveaux des pixels par leur graphe de voisinage. Ce dernier les fait appartenir `a la classe des mod`eles graphiques, dont l' ...
Arxiv preprint arXiv:0705.1613, 2007
When the number of variables $p$ is larger than the sample size $n$ of a dataset generated from a... more When the number of variables $p$ is larger than the sample size $n$ of a dataset generated from a Gaussian Graphical Model, the maximum likelihood estimation of the precision matrix does not exist. To circumvent this difficulty, in \cite{WiBu}, the authors assume a ...
The question in this paper is whether R&D efforts affect education performance in small classes. ... more The question in this paper is whether R&D efforts affect education performance in small classes. Merging two datasets collected from the PISA studies and the World Development Indicators and using Learning Bayesian Networks, we prove the existence of a statistical causal relationship between investment in R&D of a country and its education performance (PISA scores). We also prove that the effect of R&D on Education is long term as a country has to invest in R&D at least 10 years before beginning to improve the level of young pupils.

In this paper, we aim to estimate the excess of mortality associated with the severe acute respir... more In this paper, we aim to estimate the excess of mortality associated with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2/COVID-19) pandemic. For this analysis, we merged population data and deaths number provided by the World Bank database and the projections of COVID-19 deaths of 151 countries published by the Institute for Health Metrics and Evaluation (IHME). These projections are computed using a deterministic SEIR model according to the effects of non-pharmaceutical interventions. We had predicted on 12/31/2020 the new death crude rate (DCR) associated with COVID-19, its growth rate, the share of this cause of deaths in the yearly total number of deaths, and in the daily number of deaths on 12/31/2020. Excess of mortality (vs baseline growth of DCR) is varying markedly across countries. Regardless of the Scenario, Peru would be the country with the highest increase in its DCR. Many European countries, like Belgium, Italy, United Kingdom would also know a significant increase in their DCR. Tunisia would have the highest in Africa regardless of the scenario. Since the data used for this paper is dynamic and regularly updated, we had built a Shiny web application to keep the results of this paper up to date.

In this study, denoising data was advocated in sensory analysis field to remove the existing nois... more In this study, denoising data was advocated in sensory analysis field to remove the existing noise in consumer rating data before processing to External Preference Mapping (EPM). This technique is a data visualization used to understand consumers' sensory profiles by relating their preferences towards products to external information about sensory characteristics of the perceived products. The output is a perceptual map which visualizes the optimal products and aspects that maximize consumers' likings. Hence, EPM is considered as a decision tool to support the development or improvement of products and respond to market requirements. In fact, the stability of the map is affected by the high variability of judgments that make consumer rating data very noisy. This may lead to a mismatch between products features and consumers' preferences then distorted results and wrong decisions. To remove the existing noise, the use of some filtering methods is proposed. Regularized Principal Component Analysis (RPCA) and Stein's Unbiased Risk Estimate (SURE), based respectively on hard and soft thresholding rules, were applied to consumer rating data to separate the signal from noise and maintain only useful information about the given liking scores. As a way to compare the EPM obtained from each strategy, a sampling process was conducted to randomly select samples from noisy and cleaned data, then perform their corresponding EPM. The stability of the obtained maps was evaluated using an indicator that computes and compares distances between them, both before and after denoising. The effectiveness of this methodology was evaluated by a simulation study and a potential application was shown on real dataset. Results show that recorded distances after denoising are lower than those before in almost all cases for both RPCA and SURE. However, RPCA outperforms SURE. The corresponding map is made more stable where level lines are seen smoothed and products are better located on liking zones. Hence, noise removal reduces variability in data and brings closer preferences which improves the quality of the visualized map.

We studied the relationships between sediment industrial pollutants concentrations, sediment char... more We studied the relationships between sediment industrial pollutants concentrations, sediment characteristics and the dinoflagellate cyst abundance within a coastal lagoon by investigating a total of 55 sampling stations within the Bizerte lagoon, a highly anthropized Mediterranean ecosystem. The sediment of Bizerte lagoon is characterized by a high dinocyst abundance, reaching a maximum value of 2742 cysts·g −1 of dry sediment. The investigated cyst diversity was characterized by the presence of 22 dominant dinocyst morphotypes belonging to 11 genera. Two dinoflagellate species dominated the assemblage: Alexandrium pseudogonyaulax and Protoperidinium claudicans, representing 29 to 89% and 5 to 38% of the total cyst abundance, respectively, depending on the station. Seven morphotypes belonging to potentially toxic species were detected, including Alexandrium minutum, A. pseudogonyaulax, Alexandrium catenella/tamarense species complex, Lingulodinium polyedrum, Gonyaulax cf. spinifera complex, Prorocentrum micans and Protoceratium reticulatum. Pearson correlation values showed a positive correlation (α = 0.05) between cyst abundance and both water content and fine silt sediment content. Clustering revealed that the highest abundance of cysts corresponds to stations presenting the higher amounts of heavy metals. The simultaneous autoregressive model (SAM) highlighted a significant

We studied the relationships between sediment industrial pollutants concentrations, sediment char... more We studied the relationships between sediment industrial pollutants concentrations, sediment characteristics and the dinoflagellate cyst abundance within a coastal lagoon by investigating a total of 55 sampling stations within the Bizerte lagoon, a highly anthropized Mediterranean ecosystem. The sediment of Bizerte lagoon is characterized by a high dinocyst abundance, reaching a maximum value of 2742 cysts·g −1 of dry sediment. The investigated cyst diversity was characterized by the presence of 22 dominant dinocyst morphotypes belonging to 11 genera. Two dinoflagellate species dominated the assemblage: Alexandrium pseudogonyaulax and Protoperidinium claudicans, representing 29 to 89% and 5 to 38% of the total cyst abundance, respectively, depending on the station. Seven morphotypes belonging to potentially toxic species were detected, including Alexandrium minutum, A. pseudogonyaulax, Alexandrium catenella/tamarense species complex, Lingulodinium polyedrum, Gonyaulax cf. spinifera complex, Prorocentrum micans and Protoceratium reticulatum. Pearson correlation values showed a positive correlation (α = 0.05) between cyst abundance and both water content and fine silt sediment content. Clustering revealed that the highest abundance of cysts corresponds to stations presenting the higher amounts of heavy metals. The simultaneous autoregressive model (SAM) highlighted a significant

Application of Multi-SOM clustering approach to macrophage gene expression analysis
Infection, Genetics and Evolution, 2009
The production of increasingly reliable and accessible gene expression data has stimulated the de... more The production of increasingly reliable and accessible gene expression data has stimulated the development of computational tools to interpret such data and to organize them efficiently. The clustering techniques are largely recognized as useful exploratory tools for gene expression data analysis. Genes that show similar expression patterns over a wide range of experimental conditions can be clustered together. This relies on the hypothesis that genes that belong to the same cluster are coregulated and involved in related functions. Nevertheless, clustering algorithms still show limits, particularly for the estimation of the number of clusters and the interpretation of hierarchical dendrogram, which may significantly influence the outputs of the analysis process. We propose here a multi level SOM based clustering algorithm named Multi-SOM. Through the use of clustering validity indices, Multi-SOM overcomes the problem of the estimation of clusters number. To test the validity of the proposed clustering algorithm, we first tested it on supervised training data sets. Results were evaluated by computing the number of misclassified samples. We have then used Multi-SOM for the analysis of macrophage gene expression data generated in vitro from the same individual blood infected with 5 different pathogens. This analysis led to the identification of sets of tightly coregulated genes across different pathogens. Gene Ontology tools were then used to estimate the biological significance of the clustering, which showed that the obtained clusters are coherent and biologically significant.

Gender wage discrimination is a reality in the Tunisian labor market although little research has... more Gender wage discrimination is a reality in the Tunisian labor market although little research has, so far, been carried on the phenomenon. We think greater involvement of specialists should contribute to identify this phenomenon and, thus, ensure policy measures to be both legitimate and effective to help reduce it. This paper aimed to estimate the dimension of gender wage discrimination using Oaxaca-Blinder decomposition, a technique that allows us to isolate wage gap due to different characteristics, from wage gap due to different treatment for observationally equivalent characteristics,The latter is always assimilated to discrimination against women on the basis of gender and other omitted variables which cannot be observed. The results indicate that discrimination accounts for a much higher percentage of the gender monthly wage differential than do differences in the characteristics between men and women.It is especially the underestimation of female characteristics on the labor market.
Uploads
Videos by Dhafer Malouche
Papers by Dhafer Malouche