Browsing
     by title


0-9 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

or enter first few letters:   
OK
Full Text
Peer Reviewed
See detailA MACHINE-LEARNING CLASSIFIER FOR EPISODIC MIGRAINE BASED ON VISUAL EVOKED GAMMA BAND ACTIVITY
D'Ostilio, Kevin ULg; Lisicki Martinez, Marco ULg; SCHOENEN, Jean ULg et al

in Cephalalgia : An International Journal of Headache (2016), 36(1S), 56

Introduction: Objective and reliable biomarkers of migraine may be of interest for diagnosis and research purposes. Neuroimaging-based machine-learning classifiers are promising but hampered by ... [more ▼]

Introduction: Objective and reliable biomarkers of migraine may be of interest for diagnosis and research purposes. Neuroimaging-based machine-learning classifiers are promising but hampered by availability and cost issues. Conversely, evoked potential are of easy access and affordable. They have provided increasing evidence that sensory information processing is impaired in migraine. We have used gamma band oscillations (GBOs) of visual evoked potentials (VEPs) to compute a machine-learning neural network classifier in episodic migraine. Materials and methods: We analyzed GBOs from VEPs (6x100 responses). Recordings were performed in two matched samples: a training sample composed of 43 migraine patients (EM) and 20 healthy volunteers (HV) and a validating sample of 18 EM and 10 HV. A logistic regression model of the training sample was performed to evaluate the relevance of the predictor variables. Ten neural networks were automatically generated based on late component frequency, n3-p4 and p4-n4 slopes, 1st block n1-p2 amplitude and age. Results: The logistic regression model of the training sample reached a significant classification rate of 79% (EM: 88%; HV: 60%, p¼0.002). The best neural network was able to classify the groups with an accuracy of 73% in the training phase and 89% in the subsequent validation (success rate HV: 90%; EM: 88%). The mean global accuracy within the training and validating samples were 69% (63–78%) and 84% (82–89%). Conclusions: This machine-learning neural network classifier based on visual GBOs provides an accurate and costefficient tool for objective migraine diagnosis. Further training and validation studies with new cohorts are warranted [less ▲]

Detailed reference viewed: 31 (2 ULg)
Full Text
Peer Reviewed
See detailMachine-learnt versus analytical models of TCP throughput
El Khayat, Ibtissam; Geurts, Pierre ULg; Leduc, Guy ULg

in Computer Networks (2007), 51(10), 2631-2644

We first study the accuracy of two well-known analytical models of the average throughput of long-term TCP flows, namely the so-called SQRT and PFTK models, and show that these models are far from being ... [more ▼]

We first study the accuracy of two well-known analytical models of the average throughput of long-term TCP flows, namely the so-called SQRT and PFTK models, and show that these models are far from being accurate in general. Our simulations, based on a large set of long-term TCP sessions, show that 70% of their predictions exceed the boundaries of TCP-Friendliness, thus questioning their use in the design of new TCP-Friendly transport protocols. We then investigate the reasons of this inaccuracy, and show that it is largely due to the lack of discrimination between the two packet loss detection methods used by TCP, namely by triple duplicate acknowledgements or by timeout: expirations. We then apply various machine learning techniques to infer new models of the average TCP throughput. We show that they are more accurate than the SQRT and PFTK models, even without the above discrimination, and are further improved when we allow the machine-learnt models to distinguish the two loss detection techniques. Although our models are not analytical formulas, they can be plugged in transport protocols to make them TCP-Friendly. Our results also suggest that analytical models of the TCP throughput should certainly benefit from the incorporation of the timeout loss rate. (C) 2006 Elsevier B.V. All rights reserved. [less ▲]

Detailed reference viewed: 67 (3 ULg)
Full Text
See detailMachines à concepts
Durand, Pascal ULg

Article for general public (1996)

Detailed reference viewed: 10 (1 ULg)
Full Text
See detailMachines et Systèmes Thermiques
Lebrun, Jean ULg; Lemort, Vincent ULg

Learning material (2007)

Detailed reference viewed: 176 (17 ULg)
Full Text
See detailLes machines rotatives : cours de construction des machines
Hanocq, Charles ULg

Book published by Ch. Béranger - 2e édition (1935)

Detailed reference viewed: 16 (1 ULg)
Full Text
See detailLes machinimas : entre jeux et vidéos. Vers une poétique du détournement vidéoludique
Barnabé, Fanny ULg

in Culture Numérique (2015, October 10)

Cet article a pour objectif d’ébaucher la construction d’une théorie du détournement de jeux vidéo par les joueurs en passant, non pas par une perspective sociologique ou anthropologique, mais par ... [more ▼]

Cet article a pour objectif d’ébaucher la construction d’une théorie du détournement de jeux vidéo par les joueurs en passant, non pas par une perspective sociologique ou anthropologique, mais par l’analyse formelle des œuvres dérivées. Pour ce faire, le présent travail prend pour objet une pratique spécifique de détournement, la réalisation de machinimas, et l’aborde à travers un corpus illustratif de vidéos ayant pour point commun d’être issues des jeux Halo. L’analyse détaillée de quatre exemples prototypiques permettra de démontrer, d’une part, l’intime continuité qui existe entre la production de détournements et la pratique même du jeu (continuité qui rend donc artificielle toute tentative de démarcation stricte) et, d’autre part, la nature duelle du détournement de jeux vidéo (due à l’existence de deux aspects à détourner : le game et le play). La mise en parallèle de ces réflexions aboutira, enfin, dans la formalisation d’un modèle cartographiant les multiples formes de détournements. [less ▲]

Detailed reference viewed: 81 (4 ULg)
Full Text
Peer Reviewed
See detailMachining process simulation using Samcef superelement
Masset, Luc ULg; Debongnie, Jean-François ULg; Marty, Audrey

Conference (2005, February)

In this paper, we present a new simulation tool for process engineers. During process design phases, several aspects of machining have to be taken into account. Classical CAD/CAM suites still lack some ... [more ▼]

In this paper, we present a new simulation tool for process engineers. During process design phases, several aspects of machining have to be taken into account. Classical CAD/CAM suites still lack some crucial issues. The goal of the developed tool is to predict the geometric errors of machined surfaces. For classical applications of the automotive domain, form errors are mainly due to the machined part and clamping system flexibility. They are modeled thanks to the FE method. The major peculiarity of the adopted model is to apply numerous load cases. To achieve a low computational coast, we have combined the SAMCEF superelement feature and a specific code to solve the reduced system. This original scheme allows solving efficiently large industrial applications. [less ▲]

Detailed reference viewed: 187 (12 ULg)
Full Text
Peer Reviewed
See detailMachining processes simulation : specific finite element aspects
Masset, Luc ULg; Debongnie, Jean-François ULg

in Journal of Computational & Applied Mathematics (2004, July 01), 168(1-2), 309-320

The paper presents a simulation tool designed to predict form errors of part surfaces obtained by face milling and turning processes. For these operations, the form error is often due to the flexibility ... [more ▼]

The paper presents a simulation tool designed to predict form errors of part surfaces obtained by face milling and turning processes. For these operations, the form error is often due to the flexibility of the workpiece and its supports. The finite element method is adopted to model the part geometry and to compute its deformations. Numerous load cases are required to obtain the form error so that classical resolution methods prove to be inefficient (CPU time, memory and disk space). The paper mainly focuses on the special computation scheme adopted in order to improve the resolution of such an atypical problem. [less ▲]

Detailed reference viewed: 95 (31 ULg)
Full Text
Peer Reviewed
See detailMachteloos tegen de kamikazeregering 2 JAAR LINKSE OPPOSITIE
Verjans, Pierre ULg

Article for general public (2016)

Balans van 2 jaar linkse oppositie in Franstalig België. De politieke puzzel is er een stuk complexer dan in Vlaanderen, met cdH en PS regionaal aan de macht, wat op haar beurt de linkse oppositie tegen ... [more ▼]

Balans van 2 jaar linkse oppositie in Franstalig België. De politieke puzzel is er een stuk complexer dan in Vlaanderen, met cdH en PS regionaal aan de macht, wat op haar beurt de linkse oppositie tegen de regering-Michel bemoeilijkt. Die oppositie voelt zich, bijna halfweg de legislatuur, erg machteloos tegen de zogenaamde kamikazeregering gedomineerd door drie Vlaamse partijen. [less ▲]

Detailed reference viewed: 19 (2 ULg)
Full Text
See detailMacro Minitab pour l'analyse discriminante logistique
Palm, Rodolphe ULg

Software (2008)

Detailed reference viewed: 28 (3 ULg)
Full Text
Peer Reviewed
See detailMacro- and near-mesoporous monoliths by medium internal phase emulsion polymerization: a systematic study
Mathieu, Kevin ULg; Jérôme, Christine ULg; Debuigne, Antoine ULg

in Polymer (2016), 99

The synthesis of a series of poly(ethylene oxide)-b-polystyrene copolymers with different block lengths was performed by radical addition fragmentation chain transfer. These amphiphilic copolymers were ... [more ▼]

The synthesis of a series of poly(ethylene oxide)-b-polystyrene copolymers with different block lengths was performed by radical addition fragmentation chain transfer. These amphiphilic copolymers were tested as stabilizers for water-in-oil medium internal phase emulsion (MIPE) templating polymerization and the formation of polyMIPEs with controlled morphology. Aside from the structure of the emulsion stabilizer, several parameters susceptible to influence the size of the cavities and the interconnectivity of the porous monoliths were probed including the choice of the comonomers, treatment of the emulsion by ultrasound, the use of controlled radical polymerization method for the network formation as well as interfacial initiation. Interconnected cellular monoliths were produced. The polymerization of the ultrasonicated water-in-ethylhexylacrylate/divinylbenzene MIPE notably led to near-mesoporous open-cell material. Mechanical properties and specific surface areas of the polyMIPEs were also investigated and discussed. [less ▲]

Detailed reference viewed: 44 (19 ULg)
Full Text
Peer Reviewed
See detailMacro-economic factors influencing the architectural business model shift in the pharmaceutical industry
Dierks, Raphaela Marie Louisa ULg; Bruyère, Olivier ULg; Reginster, Jean-Yves ULg et al

in Expert Review of Pharmacoeconomics & Outcomes Research (2016), 16(5), 571-8

Introduction: Technological innovations, new regulations, increasing costs of drug productions and new demands are only few key drivers of a projected alternation in the pharmaceutical industry and its ... [more ▼]

Introduction: Technological innovations, new regulations, increasing costs of drug productions and new demands are only few key drivers of a projected alternation in the pharmaceutical industry and its underlying business model. Despite significant articles in the press, there is little medical and economic evidence on the rationale driving the transformation from the traditional Research & Development (R&D) model to new alternative business models, such as Mergers & Acquisitions (M&A). The purpose of this review is to understand the macro-economic factors responsible for the revolution of the pharmaceutical business model, by understanding the needs, desires, new behavioral patterns, but also challenges within the industry. Considering these changes will be the key to possessing a competitive advantage over the market players. Areas covered: Existing literature on new macro-economic factors changing the pharmaceutical landscape has been reviewed to present a clear image of the current market environment. Expert commentary: Literature and media sources show that pharmaceutical companies are facing an architectural alteration in their business model, with more M&A deals and collaborations headlining the papers. Although Q1 2016 did show a major slowdown in M&A deals by volume since 2013 (with deal cancellations of Pfizer and Allergan, or the downfall of Valeant) pharmaceutical analysts remain confident that this shortfall was a consequence of the equity market volatility. It seems likely that the shift to an M&A model will become apparent during the remainder of 2016, with deal announcements such as Abbott Laboratories, AbbVie and Sanofi worth USD 45billion [1] showing the appetite of big pharma companies to shift from the fully vertical integrated business model to more horizontal business models. [less ▲]

Detailed reference viewed: 75 (18 ULg)
Full Text
Peer Reviewed
See detailA MACRO-ELEMENT FOR THE NONLINEAR ANALYSIS OF DEEP BEAMS BASED ON A THREE-PARAMETER KINEMATIC MODEL
Liu, Jian ULg; Mihaylov, Boyan ULg

Conference (2016, November 22)

This paper presents a computationally-effective macro element for predicting the complete shear response of RC deep beams. The element stems from a three-parameter kinematic theory (3PKT) for the shear ... [more ▼]

This paper presents a computationally-effective macro element for predicting the complete shear response of RC deep beams. The element stems from a three-parameter kinematic theory (3PKT) for the shear strength and deformation capacity of deep beams under double curvature. The macro element consists of two rotational springs and one transverse spring to simulate the nonlinear flexural and shear behaviour of deep shear spans. The element is applied to two deep beams tested to shear failure under three-point bending. The results are used to study the effect of unsymmetrical shear failures on the pre- and post-peak behaviour of nominally symmetrical members. It is found that the post-peak response is sensitive to the unloading path of the shear resisting mechanisms as well as to the global roughness of the critical diagonal cracks. Based on these analyses, modifications to the constitutive relationships of the shear mechanisms are proposed, and the macro element is applied to a continuous deep beam. It is shown that the model adequately captures the complete behaviour of the beam, including the redistribution of forces following the shear failure in the critical shear span. [less ▲]

Detailed reference viewed: 46 (19 ULg)
Full Text
Peer Reviewed
See detailA Macro-Element Formulation and Solution Procedure for Shear Analysis of RC Deep Beams
Liu, Jian ULg; Mihaylov, Boyan ULg

in the 11th fib International PhD Symposium in Civil Engineering (2016, August 30)

This paper presents a computationally-effective macro element capable of predicting the complete shear response of RC deep beams in analogy with beam elements for slender beams. The element stems from a ... [more ▼]

This paper presents a computationally-effective macro element capable of predicting the complete shear response of RC deep beams in analogy with beam elements for slender beams. The element stems from a three-parameter kinematic theory (3PKT) for the shear strength and deformation capacity of deep beams subjected to double curvature. The macro element is composed of two rotational springs and one transverse spring to simulate the nonlinear flexural and shear behaviour of the shear span. The element has two nodes (end sections) with two DOFs per node, which enables it to be easily connected to other elements of the same or different types to model structures with deep beams. The numerical implementation of the element is based on the secant stiffness approach which provides robust convergence properties. Validation studies are performed by using thirteen tests from the literature showing excellent results for the complete shear response. The obtained peak load experimental-to-predicted ratios have an average of 1.10 with a coefficient of variation of 14.2%, while these values for the mid-span deflection are 1.16 and 17.0%. [less ▲]

Detailed reference viewed: 71 (28 ULg)
Full Text
Peer Reviewed
See detailMacro-Kinematic Approach for Shear Behaviour of Short Coupling Beams with Conventional Reinforcement
Mihaylov, Boyan ULg; Franssen, Renaud ULg

Conference (2017, June 13)

Short coupling beams in wall structures work predominantly in shear and develop complex deformation patterns. For this reason they cannot be modelled based on the classical plane-sections-remain-plane ... [more ▼]

Short coupling beams in wall structures work predominantly in shear and develop complex deformation patterns. For this reason they cannot be modelled based on the classical plane-sections-remain-plane hypothesis, and are typically designed with strut-and-tie models. However, because strut-and-tie models are inherently conservative, they can result in very large amounts of shear reinforcement (stirrups), and therefore significant construction difficulties. In addition, strut-and-tie models do not provide information about the deformation capacity of coupling beams, which is important for performance-based seismic design. To address these challenges, this paper discusses a three-parameter kinematic theory (3PKT) for the shear strength and deformation patterns of short coupling beams. The 3PKT approach is situated between simple and conservative strut-and-tie models and complex non-linear finite element (FE) models. While FE models use a large number of degrees of freedom (DOFs) to describe the deformation patterns in coupling beams, the 3PKT method is based on a kinematic model with only three DOFs. The paper presents the formulation of the model and its validation with tests. [less ▲]

Detailed reference viewed: 25 (7 ULg)