Institut de Recherche en Informatique de Toulouse
Recent publications
The connection between humans and digital technologies has been documented extensively in the past decades but needs to be evaluated through the current global pandemic. Artificial Intelligence(AI), with its two strands, Machine Learning (ML) and Semantic Reasoning, has proven to be a great solution to provide efficient ways to prevent, diagnose and limit the spread of COVID-19. IoT solutions have been widely proposed for COVID-19 disease monitoring, infection geolocation, and social applications. In this paper, we investigate the usage of the three technologies for handling the COVID-19 pandemic. For this purpose, we surveyed the existing ML applications and algorithms proposed during the pandemic to detect COVID-19 disease using symptom factors and image processing. The survey includes existing approaches including semantic technologies and IoT systems for COVID-19. Based on the survey result, we classified the main challenges and the solutions that could solve them. The study proposes a conceptual framework for pandemic management and discusses challenges and trends for future research.
Background and aims Maternal diet plays a key role in preventing or contributing to the development of chronic diseases, such as obesity, allergy, and brain disorders. Supplementation of maternal diet with prebiotics has been shown to reduce the risk of food allergies and affect the intestinal permeability in offspring later in life. However, its role in modulating the development of other intestinal disorders, such as colitis, remains unknown. Therefore, we investigated the effects of prebiotic supplementation in pregnant mice on the occurrence of colitis in their offspring. Materials and methods Offspring from mothers, who were administered prebiotic galacto-oligosaccharides and inulin during gestation or fed a control diet, were subjected to three cycles of dextran sulphate sodium (DSS) treatment to induce chronic colitis, and their intestinal function and disease activity were evaluated. Colonic remodelling, gut microbiota composition, and lipidomic and transcriptomic profiles were also assessed. Results DSS-treated offspring from prebiotic-fed mothers presented a higher disease score, increased weight loss, and increased faecal humidity than those from standard diet-fed mothers. DSS-treated offspring from prebiotic-fed mothers also showed increased number of colonic mucosal lymphocytes and macrophages than the control group, associated with the increased colonic concentrations of resolvin D5, protectin DX, and 14-hydroxydocosahexaenoic acid, and modulation of colonic gene expression. In addition, maternal prebiotic supplementation induced an overabundance of eight bacterial families and a decrease in the butyrate caecal concentration in DSS-treated offspring. Conclusion Maternal prebiotic exposure modified the microbiota composition and function, lipid content, and transcriptome of the colon of the offspring. These modifications did not protect against colitis, but rather sensitised the mice to colitis development.
The transition from the 20th- to 21st-century education is a vital and complex process for Higher Education Institutions (HEIs). It requires constant updating of their strategic plan, ensuring all internal competencies necessary for improving and adapting their teaching and learning processes are in place. This paper presents the design and evaluation of the PROF-XXI tool, a web-based dashboard designed to help HEIs make such transformation sustainable and support their Teaching and Learning Centres (TLCs) in decision making about teaching and learning strategies. The tool is based on the PROF-XXI competency framework, which offers a holistic perspective on the competencies that TLCs should develop to support deep and sustainable transformations. Through a structured interview carried out with 17 participants from 8 HEIs, this study evaluates the extent to which the proposed tool assists users with identifying the required competencies and monitoring their evolution. The results show the value of the tool to improve and monitor the institutions’ innovative teaching and learning competencies, the most appropriate indicators for this purpose and suggestions on adaptations to the existing features of the tool.
The practical Byzantine Fault Tolerance (PBFT) is a classical consensus algorithm that has been widely applied in an alliance blockchain system to make all nodes agree to certain transactions under the assumption that the proportion of Byzantine nodes is no more than 1/3. It is prevalent due to its performance, simplicity, and claimed correctness. However, any vulnerability of the consensus algorithm can lead to a significant loss in finance because no one can change the transaction results after execution. This paper proposes a formal development method of the PBFT algorithm by horizontal refinement in Event-B, which allows us to manage the complexity of the proof process by factoring the proof of correctness into several refinement steps. During the development of PBFT, we have specified the core mechanism like parameterized message types, primary node change, and water-mark interval. Furthermore, we present a mechanical verification of the safety and liveness properties of the model in Rodin, which can be partially and widely used to check the blockchain consensus algorithm vulnerability using a refinement tree of algorithms.
Speckle has a considerable impact on medical ultrasound (US) imaging due to its its intrinsic random nature and spatially correlated behavior that severely reduces image contrast. In this paper, leveraging the quantum many-body theory, we propose a deep-learning architecture recasting a baseline denoising algorithm for adaptive contrast enhancement of US images. The proposed deep neural network integrates quantum mechanical concepts based on Schrödinger equations which makes our model robust for US image restoration. The potential of the proposed deep unfolded network is illustrated on simulated data and clinical cardiac US images. Both results show enhanced images with improved contrast and resolution while preserving underlying structures and significantly reducing the speckle noise.
The presence of the dialect in the Arabic texts made Arabic sentiment analysis (ASA) a challenging issue, owing to it usually does not follow specific rules in writing systems, especially Tunisian Dialectical (TD) which presents an undertaking challenge due to its complexity, ambiguity, the morphological richness of the language, the absence of contextual information, the code-switching (CS) and mostly the multilingualism phenomena in textual productions. Recently, deep learning models have clearly demonstrated a great success in the field of sentiment analysis (SA). Although, the state-of-the-art accuracy for dialectical sentiment analysis (DSA) still needs improvements regarding contextual information and implicit sentiment expressed in different real cases. To address this challenge, we propose, an efficient Bidirectional LSTM network preceded by a preprocessing stage in order to enhance Tunisian SA, by applying Forward-Backward encapsulate contextual information from multilingual feature sequences. To evaluate our model, and due to the lack of publicly available multilingual resources associated with the TD, we collect different datasets available with different variants of TD to create our own multilingual corpus for sentiment classification. The experimental results based on the evaluation standards “Accuracy”, “Recall” and “F1-score” demonstrate that our model achieves significant improvements over the state-of-art deep learning models and the baseline traditional machine learning methods.
Self-regulated learning (SRL) is a crucial higher-order skill required by learners of the 21st century, who will need to become lifelong learners to adapt to the continually changing environments. Literature provides examples of tools for scaffolding SRL in online environments. In this article, we provide the state-of-the-art concerning tools that support SRL in terms of theoretical models underpinning development, supported SRL processes, tool functionalities, used data and visualizations. We reviewed 42 articles published between 2008 and 2020, including information from 25 tools designed to support SRL. Our findings indicate that: 1) many of the studies do not explicitly specify the SRL theoretical model used to guide the design process of the tool; 2) goal setting, monitoring, and self-evaluation are the most prevalent SRL processes supported through functionalities, such as content navigation, user input forms, collaboration features, and recommendations; 3) the relationship between tool functionalities and SRL processes are rarely described; and 4) few tools assess the impact on learners’ SRL process and learning performance. Finally, we highlight some lessons learned that might contribute to implementing future tools that support learners’ SRL processes.
The definition of Megalithism is found in specialized and academic works, on Archaeology and Prehistory, as well as in generalist dictionaries, glossaries and encyclopedias. However we are not aware of formal definitions specifically to this particular domain. This paper presents an under development proposal of Megalithism Knowledge Representation that relies on CIDOC CRM to represent the monuments and concepts of European Megalithism. It includes the granularity required to represent this form of architecture: from composite parts to single elements, such as standing stones. A structured way for representing such definitions is required in order to guide future knowledge extraction from Megalithism reports. KeywordsMegalithismKnowledge representationCIDOC CRM
Capacity building for Learning Analytics (LA) in Higher Education Institutions requires the coordination of organizational aspects and infrastructure development. This also depends on the organizational maturity of the institution and its leadership regarding LA adoption. LA capacity building can follow two approaches: (1) top-down, led by institutional managers; and (2) bottom-up, led by ground-level staff. This article studies two LA initiatives of each type conducted in the same institution to compare the deployment of organizational processes and infrastructure. The lessons learned that were captured from each approach are shared to inform other universities in Latin America on developing LA capabilities.
A digraph D is singly connected if for all ordered pairs of vertices u,v∈V(D), there is at most one path in D from u to v. In this paper, we study the Singly Connected Vertex Deletion (SCVD) problem: Given an n-vertex digraph D and a positive integer k, does there exist a set S⊆V(D) such that |S|≤k and D−S is singly connected? This problem may be seen as a directed counterpart of the (Undirected) Feedback Vertex Set problem, as an undirected graph is singly connected if and only if it is acyclic. SCVD is known to be NP-hard on general digraphs. We study the complexity of SCVD on various classes of digraphs such as tournaments, and various generalisations of tournaments such as digraphs of bounded independence number, in- and out-tournaments and local tournaments. We show that unlike the Feedback Vertex Set on Tournaments (FVST) problem, SCVD is polynomial-time solvable on tournaments. In addition, we show that SCVD is polynomial-time solvable on digraphs of bounded independence number, and on the class of acyclic local tournaments. We also study the parameterized complexity of SCVD, with k as the parameter, on the class of in-tournaments. And we show that on in-tournaments, SCVD admits a fixed-parameter tractable algorithm and a quadratic vertex kernel. We also show that on the class of local tournaments, which is a sub-class of in-tournaments, SCVD admits a linear vertex kernel.
In this paper, we focus on a recommendation approach for collaborative decision-making. This recommendation is possible by designing an ontology in a first step. The construction of this ontology is based on a state of the art on collaborative decision-making, on ontology engineering and on collaboration engineering. An eight-step ontology development methodology was adopted and implemented to design the ontology. A corpus made up of one hundred and seventy-seven (177) documents was the starting point for the extraction of terms from the ontology and the Unified Modeling Language (UML) language is used as a description language of the ontology. This ontology is the starting point for a facilitation support system in a collaborative decision-making process. In a second step, we defined rules, based on our expertise on collaborative decision-making, used to guide a group of decision maker or a facilitator.
Multifractal analysis has become a reference tool for signal and image processing. Grounded in the quantification of local regularity fluctuations, it has proven useful in an increasing range of applications, yet so far involving only univariate data (scalar valued time series or single channel images). Recently the theoretical ground for multivariate multifractal analysis has been devised, showing potential for quantifying transient higher-order dependence beyond linear correlation among collections of data. However, the accurate estimation of the parameters associated with a multivariate multifractal model remains challenging, especially for small sample size data. This work studies an original Bayesian framework for multivariate multifractal estimation, combining a novel and generic multivariate statistical model, a Whittle-based likelihood approximation and a data augmentation strategy allowing parameter separability. This careful design enables efficient estimation procedures to be constructed for two relevant choices of priors using a Gibbs sampling strategy. Monte Carlo simulations, conducted on synthetic multivariate signals and images with various sample sizes and multifractal parameter settings, demonstrate significant performance improvements over the state of the art, at only moderately larger computational cost. Moreover, we show the relevance of the proposed framework for real-world data modeling in the important application of drowsiness detection from multichannel physiological signals.
Institution pages aggregate content on ResearchGate related to an institution. The members listed on this page have self-identified as being affiliated with this institution. Publications listed on this page were identified by our algorithms as relating to this institution. This page was not created or approved by the institution. If you represent an institution and have questions about these pages or wish to report inaccurate content, you can contact us here.
328 members
Malik Muhammad Saad Missen
  • Institut de Recherche en Informatique de Toulouse (IRIT)
Julien Pinquier
  • Structuration, Analysis, MOdeling of Video and Audio Team (SAMoVA)
Nathalie Aussenac-Gilles
  • Methodes et Ingénierie des Langues, des Ontologies et du DIscours Team (MELODI)
Thierry Val
  • Réseaux - Mobiles - Embarqués - Sans fil - Satellites (RMESS)
Gilles Hubert
  • University of Toulouse
118 route de Narbonne, 31062 cedex 9, Toulouse, Midi-Pyrenees, France