Time filter

Source Type

Ben Messaoud M.,Institute Superieur Of Gestion Tunis | Ben Messaoud M.,CNRS Nantes Atlantic Computer Science Laboratory | Leray P.,CNRS Nantes Atlantic Computer Science Laboratory | Ben Amor N.,Institute Superieur Of Gestion Tunis
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2011

Learning Causal Bayesian Networks (CBNs) is a new line of research in the machine learning field. Within the existing works in this direction [8, 12, 13], few of them have taken into account the gain that can be expected when integrating additional knowledge during the learning process. In this paper, we present a new serendipitous strategy for learning CBNs using prior knowledge extracted from ontologies. The integration of such domain's semantic information can be very useful to reveal new causal relations and provide the necessary knowledge to anticipate the optimal choice of experimentations. Our strategy also supports the evolving character of the semantic background by reusing the causal discoveries in order to enrich the domain ontologies. © 2011 Springer-Verlag Berlin Heidelberg.


Ben Messaoud M.,CNRS Nantes Atlantic Computer Science Laboratory | Ben Messaoud M.,Institute Superieur Of Gestion Tunis | Leray P.,CNRS Nantes Atlantic Computer Science Laboratory | Ben Amor N.,Institute Superieur Of Gestion Tunis
Knowledge-Based Systems | Year: 2015

Within the last years, probabilistic causality has become a very active research topic in artificial intelligence and statistics communities. Due to its high impact in various applications involving reasoning tasks, machine learning researchers have proposed a number of techniques to learn Causal Bayesian Networks. Within the existing works in this direction, few studies have explicitly considered the role that decisional guidance might play to alternate between observational and experimental data processing. In this paper, we go further by introducing a serendipitous strategy to elucidate semantic background knowledge provided by the domain ontology to learn the causal structure of Bayesian Networks. We also complement our contribution with an enrichment process by which it will be possible to reuse these causal discoveries, support the evolving character of the semantic background and make an ontology evolution. Finally, the proposed method will be validated through simulations and real data analysis. © 2014 Elsevier B.V. All rights reserved.


Amor N.B.,Institute Superieur Of Gestion Tunis | Dubois D.,French National Center for Scientific Research | Gouider H.,Institute Superieur Of Gestion Tunis | Prade H.,French National Center for Scientific Research
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2015

The paper discusses the use of product-based possibilistic networks for representing conditional preference statements on discrete variables. The approach uses non-instantiated possibility weights to define conditional preference tables. Moreover, additional information about the relative strengths of symbolic weights can be taken into account. It yields a partial preference order among possible choices corresponding to a symmetric form of Pareto ordering. In the case of Boolean variables, this partial ordering coincides with the inclusion between the sets of preference statements that are violated. Furthermore, this graphical model has two logical counterparts in terms of possibilistic logic and penalty logic. The flexibility and the representational power of the approach are stressed. Besides, algorithms for handling optimization and dominance queries are provided. © Springer International Publishing Switzerland 2015.


Ayachi R.,Institute Superieur Of Gestion Tunis | Ayachi R.,University of Artois | Ben Amor N.,Institute Superieur Of Gestion Tunis | Benferhat S.,University of Artois
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2011

Qualitative causal possibilistic networks are important tools for handling uncertain information in the possibility theory framework. Despite their importance, no compilation has been performed to ensure causal reasoning in possibility theory framework. This paper proposes two compilation-based inference algorithms for min-based possibilistic causal networks. The first is a possibilistic adaptation of the probabilistic inference method [8] and the second is a purely possibilistic approach. Both of them are based on an encoding of the network into a propositional theory and a compilation of this output in order to efficiently compute the effect of both observations and interventions, while adopting a mutilation strategy. © 2011 Springer-Verlag Berlin Heidelberg.


Ayachi R.,Institute Superieur Of Gestion Tunis | Ayachi R.,University of Artois | Ben Amor N.,Institute Superieur Of Gestion Tunis | Benferhat S.,University of Artois
Fuzzy Sets and Systems | Year: 2014

Qualitative possibilistic causal networks are important tools for handling uncertain information in the possibility theory framework. Contrary to possibilistic networks (Ayachi et al., 2011 [2]), the compilation principle has not been exploited to ensure causal reasoning in the possibility theory framework. This paper proposes mutilated-based inference approaches and augmented-based inference approaches for qualitative possibilistic causal networks using two compilation methods. The first one is a possibilistic adaptation of the probabilistic inference approach (Darwiche, 2002 [13]) and the second is a purely possibilistic approach based on the transformation of the graphical-based representation into a logic-based one (Benferhat et al., 2002 [3]). Each of the proposed methods encodes the network or the possibilistic knowledge base into a propositional base and compiles this output in order to efficiently compute the effect of both observations and interventions. This paper also reports on a set of experimental results showing cases in which augmentation outperforms mutilation under compilation and vice versa. © 2013 Elsevier B.V. Published by Elsevier B.V. All rights reserved.


Ayachi R.,Institute Superieur Of Gestion Tunis | Ayachi R.,University of Artois | Ben Amor N.,Institute Superieur Of Gestion Tunis | Benferhat S.,University of Artois
Information Sciences | Year: 2014

Probabilistic and possibilistic networks are important tools proposed for an efficient representation and analysis of uncertain information. The inference process has been studied in depth in these graphical models. We cite in particular compilation-based inference which has recently triggered the attention of several researchers. In this paper, we are interested in comparing this inference mechanism in the probabilistic and possibilistic frameworks in order to unveil common points and differences between these two settings. In fact, we will propose a generic framework supporting both Bayesian networks and possibilistic networks (product-based and min-based ones). The proposed comparative study points out that the inference process depends on the specificity of each framework, namely in the interpretation of the handled uncertainty degrees (probability\possibility) and appropriate operators (*\min and +\max). © 2013 Elsevier Inc. All rights reserved.


Badreddine A.,Institute Superieur Of Gestion Tunis | Ben Amor N.,Institute Superieur Of Gestion Tunis
2010 ACS/IEEE International Conference on Computer Systems and Applications, AICCSA 2010 | Year: 2010

Bow tie diagrams have become popular methods in risk analysis and safety management. This tool describes graphically, in the same scheme, the whole scenario of an identified risk and its respective preventive and protective barriers. The major problem with bow tie diagrams is that they remain limited by their technical level and by their restriction to a graphical representation of different scenario without any consideration to the dynamic aspect of real systems. Recently we have proposed a new Bayesian approach to construct bow tie diagrams for risk analysis [1]. This approach learns the bow tie structure from real data and improves them by adding a new numerical component allowing us to model in a more realistic manner the system behavior. In this paper we propose to extend this approach by adding the barriers implementation in order to construct the whole bow ties. To this end we will use the numerical component, previously defined in the learning phase, and the analytic hierarchical process (AHP).


Slokom M.,Institute Superieur Of Gestion Tunis | Ayachi R.,Institute Superieur Of Gestion Tunis
2015 2nd International Conference on Computer Science, Computer Engineering, and Social Media, CSCESM 2015 | Year: 2015

Collaborative filtering approaches exploit users preferences to provide items recommendations. These preferences describing the actual state of the item are generally certain. However, in real problems we can not ignore the importance of uncertainty. In this paper, we propose a purely possibilistic collaborative filtering approach that provides a recommendation list given uncertain preferences expressed by possibility distributions. Experimental results show that the proposed approach outperforms traditional collaborative filtering algorithm. © 2015 IEEE.


Badreddine A.,Institute Superieur Of Gestion Tunis | Ben Amor N.,Institute Superieur Of Gestion Tunis
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2010

Bow tie diagrams have become popular methods in risk analysis and safety management. This tool describes graphically, in the same scheme, the whole scenario of an identified risk and its respective preventive and protective barriers. The major problem with bow tie diagrams is that they remain limited by their technical level and by their restriction to the graphical representation of different scenarios without any consideration to the dynamic aspect of real systems. This paper overcomes this weakness by proposing a new Bayesian approach to construct bow ties from real data. © 2010 Springer-Verlag.


Slimen Y.B.,Institute Superieur Of Gestion Tunis | Ayachi R.,Institute Superieur Of Gestion Tunis | Amor N.B.,Institute Superieur Of Gestion Tunis
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2013

Probability-possibility transformation is a purely mechanical transformation of probabilistic support to possibilistic support and vice versa. In this paper, we apply the most common transformations to graphical models, i.e., Bayesian into possibilistic networks. We show that existing transformations are not appropriate to transform Bayesian networks to possibilistic ones since they cannot preserve the information incorporated in joint distributions. Therefore, we propose new consitency properties, exclusively useful for graphical models transformations. © Springer International Publishing 2013.

Loading Institute Superieur Of Gestion Tunis collaborators
Loading Institute Superieur Of Gestion Tunis collaborators