Roubaix, France
Roubaix, France

Time filter

Source Type

Guettala A.E.-T.,University of Tours | Bouali F.,University of Lille2 | Guinot C.,University of Tours | Guinot C.,Biometrics and Epidemiology Unit | Venturini G.,University of Tours
Proceedings of the International Conference on Information Visualisation | Year: 2012

We deal in this paper with the problem of automating the process of choosing an appropriate visualization and its parameters in the context of visual data mining (VDM). To solve this problem, we develop a user assistant that performs 2 steps: the system starts by suggesting to users different mappings between their data and possible visualizations. This is performed with a simple but generic heuristic that can be applied to any visualization. Then, the user selects a visualization among those proposed by our assistant, and he may further improve the parameters set that defines the mapping between the visual attributes and the data attributes. For this purpose, we use an interactive genetic algorithm (IGA), which allows users to visually evaluate and adjust the mappings. We present a user evaluation that confirms the interest of our system in two tasks. © 2012 IEEE.


Bouali F.,University of Lille2 | Devaux S.,Airbus | Venturini G.,University of Tours
Visual Computer | Year: 2016

In this paper, we study the visual mining of time series, and we contribute to the study and evaluation of 3D tubular visualizations. We describe the state of the art in the visual mining of time-dependent data, and we concentrate on visualizations that use a tubular shape to represent data. After analyzing the motivations for studying such a representation, we present an extended tubular visualization. We propose new visual encodings of the time and data, new interactions for knowledge discovery, and the use of rearrangement clustering. We show how this visualization can be used in several real-world domains and that it can address large datasets. We present a comparative user study. We conclude with the advantages and the drawbacks of our method (especially the tubular shape). © 2014, Springer-Verlag Berlin Heidelberg.


Bouali F.,University of Lille2 | Bouali F.,University of Tours | Guettala A.,University of Tours | Venturini G.,University of Tours
Visual Computer | Year: 2015

We study in this work how a user can be guided to find a relevant visualization in the context of visual data mining. We present a state of the art on the user assistance in visual and interactive methods. We propose a user assistant called VizAssist, which aims at improving the existing approaches along three directions: it uses simpler computational models of the visualizations and the visual perception guidelines, in order to facilitate the integration of new visualizations and the definition of a mapping heuristic. VizAssist allows the user to provide feedback in a visual and interactive way, with the aim of improving the data to visualization mapping. This step is performed with an interactive genetic algorithm. Finally, VizAssist aims at proposing a free on-line tool (www.vizassist.fr) that respects the privacy of the user data. This assistant can be viewed as a global interface between the user and some of the many visualizations that are implemented with D3js. © 2015 Springer-Verlag Berlin Heidelberg


Liu T.,University of Tours | Bouali F.,University of Lille2 | Venturini G.,University of Tours
Distributed and Parallel Databases | Year: 2015

In this paper, we study how to visualize large amounts of multidimensional data with a radial visualization. For such a visualization, we study a multi-threaded implementation on the CPU and the GPU. We start by reviewing the approaches that have visualized the largest multidimensional datasets and we focus on the approaches that have used CPU or GPU parallelization. We consider the radial visualizations and we describe our approach (called POIViz) that uses points of interest to determine a layout of a large dataset. We detail its parallelization on the CPU and the GPU. We study the efficiency of this approach with different configurations and for large datasets. We show that it can visualize, in less than one second, millions of data with tens of dimensions, and that it can support “real-time” interactions even for large datasets. We conclude on the advantages and limits of the proposed visualization. © 2015 Springer Science+Business Media New York


Fanchini M.,University of Verona | Castagna C.,Football Training and Biomechanics Laboratory | Coutts A.J.,University of Technology, Sydney | Schena F.,University of Verona | And 2 more authors.
Journal of Sports Sciences | Year: 2014

Abstract: The aim of this study was to compare the reliability, internal responsiveness and interchangeability of the Yo-Yo intermittent recovery test level 1 (YY1), level 2 (YY2) and submaximal YY1 (YY1-sub). Twenty-four young soccer players (age 17 ± 1 years; height 177 ± 7 cm; body mass 68 ± 6 kg) completed each test five times within pre- and in-season; distances covered and heart rates (HRs) were measured. Reliability was expressed as typical error of measurement (TEM) and intraclass correlation coefficient (ICC). Internal responsiveness was determined as effect size (ES) and signal-to-noise ratio (ESTEM). Interchangeability was determined with correlation between training-induced changes. The TEM and ICC for distances in the YY1 and YY2 and for HR in YY1-sub were 7.3% and 0.78, 7.1% and 0.93 and 2.2% and 0.78, respectively. The ESs and ESTEMs were 0.9 and 1.9 for YY1, 0.4 and 1.2 for YY2 and −0.3 and −0.3 for YY1-sub. Correlations between YY1 vs. YY2 and YY1-sub were 0.56 to 0.84 and −0.36 to −0.81, respectively. Correlations between change scores in YY1 vs. YY2 were 0.29 and −0.21 vs. YY1-sub. Peak HR was higher in YY1 vs. YY2. The YY1 and YY2 showed similar reliability; however, they were not interchangeable. The YY1 was more responsive to training compared to YY2 and YY1-sub. © 2014, © 2014 Taylor & Francis.


Liu T.,University of Tours | Bouali F.,University of Tours | Bouali F.,University of Lille2 | Venturini G.,University of Tours
Proceedings of the International Conference on Information Visualisation | Year: 2015

We study in this paper the visualization of large multidimensional datasets with a focus on Open Data. Starting from our early work in which we defined a visualization based on points of interest, we improve this method in several ways with the aim of dealing with larger datasets and especially Open datasets. We propose the parallelization, using CPU and GPU, of the most costly steps of our method, like the computation of the data layout. We improve the visualization with a density rendering so as to keep the display informative for large datasets and for Open Data. We propose a layered visualization with interactions that can support several users tasks such as data filtering and labeling. We show that, even with common hardware, the performances of our approach are such that any user graphical queries can be processed in a few seconds. We detail how we were able to visualize and explore a collection of 300,000 Open datasets from the French Open Data web site. With the resulting visualization, we were able to improve our previous results. © 2015 IEEE.


Sese M.A.,University of Zaragoza | Jimenez-Pavon D.,University of Zaragoza | Gilbert C.C.,Campden and Chorleywood Food Research Association | Gonzalez-Gross M.,Technical University of Madrid | And 11 more authors.
Appetite | Year: 2012

The present study examined the associations of food behaviours and preferences with markers of insulin resistance and clustered metabolic risk factors score after controlling for potential confounders, including body fat in European adolescents. A cross-sectional study " Healthy Lifestyle in Europe by Nutrition in Adolescence Cross-Sectional Study" of 3546 European adolescents aged 12.5-17.5. years was conducted, using a complete dataset on at least glucose, insulin and " Food Choice Questionnaire" Results indicated skipping breakfast, as well as the preference of some foods such as nuts, chocolate, burgers and pizzas, soft drinks or juices, explain part of homeostasis model assessment index variance. In addition, snacking regularly during school day is associated with higher metabolic risk score in females. In conclusion, the present findings suggest that intervention studies aimed to prevent insulin resistance and metabolic risk factors in youth should focus not only in influencing food and drink preferences, but also to ensure healthy food behaviour in adolescents. The harmful consequences in the choice of certain foods or drinks and food habits can be countered with proper planning and intervention programs to prevent insulin resistance and metabolic risk factors. © 2012 Elsevier Ltd.


Fanchini M.,University of Verona | Schena F.,University of Verona | Castagna C.,Football Training and Biomechanics Laboratory | Castagna C.,University of Rome Tor Vergata | And 4 more authors.
International Journal of Sports Medicine | Year: 2015

The aim of this study was to assess the external responsiveness, construct validity and internal responsiveness of the Yo-Yo Intermittent Recovery test level 1 and its sub-maximal version in semi-professional players. Tests and friendly matches were performed during the preseason and regular season. The distance covered above 15 km·h-1 was considered as an indicator of the physical match performance. Construct validity and external responsiveness were examined by correlations between test and physical match performance (preseason and regular season) and training-induced changes. Internal responsiveness was determined as Cohen's effect size, standardized response mean and signal-to-noise ratio. The physical match performance increased after training (34.8%). The Yo-Yo Intermittent Recovery test level 1 improved after training (40.2%), showed longitudinal (r=0.69) and construct validity (r=0.73 and 0.59, preseason and regular season) and had higher internal responsiveness compared to its sub-maximal version. The heart rate at the 6th minute in the sub-maximal version did not show longitudinal (r=-0.38) and construct validity (r=0.01 and -0.06, preseason and regular season) and did not significantly change after training (-0.3%). The rate of perceived exertion decreased in the sub-maximal version (-29.8%). In conclusion, the Yo-Yo Intermittent Recovery test level 1 is valid and responsive, while the validity of its sub-maximal version is questionable. © Georg Thieme Verlag KG Stuttgart · New York.


Jimenez-Pavon D.,University of Zaragoza | Jimenez-Pavon D.,University of Granada | Jimenez-Pavon D.,Karolinska Institutet | Ruiz J.R.,Karolinska Institutet | And 12 more authors.
Pediatric Diabetes | Year: 2013

Objective: To analyze the association between objectively assessed physical activity (PA) and markers of insulin resistance (IR) in European adolescents and to examine whether the association of objectively assessed PA and markers of IR is modified by cardiorespiratory fitness (CRF). Methods: A cross-sectional study at school setting was conducted in 1053 (554 girls) adolescents (12.5-17.5 yr). Weight, height, waist circumference, total body fat, PA (accelerometry), and CRF (20-m shuttle run test) were measured. Fasting insulin and glucose concentrations were measured, and homeostasis model assessment as well as quantitative insulin sensitivity check index were computed. Results: In males, vigorous PA (VPA) was negatively associated with markers of IR after adjusting for confounders including waist circumference (all p<0.05). In females, moderate PA, moderate to vigorous PA, and average PA were negatively associated with markers of IR after adjusting for confounders (all p<0.05). Moreover, when the sample was segmented by CRF levels all the PA intensities were significantly associated with the markers of IR in females with low CRF but not in those with middle-high CRF after adjusting for confounders. Conclusions: The findings suggest that PA is negatively associated with markers of IR after adjusting for confounders including total and central body fat in both sexes, but this relationship is modified by the CRF levels being especially important in those females with low CRF. Preventive strategies should focus not only on increasing the volume of PA but also on enhancing CRF through VPA. © 2013 John Wiley & Sons A/S.


Liu T.,University of Tours | Bouali F.,University of Tours | Bouali F.,University Of Lille2 | Venturini G.,University of Tours
Revue des Nouvelles Technologies de l'Information | Year: 2013

In this paper, we propose to parallelize, on CPU and GPU, a radial-based visualization method that uses points of interests.We show that this approach may visualize in a few seconds millions of data with tens of dimensions, and we study the efficiency of the parallel approach in different configurations.

Loading University of Lille2 collaborators
Loading University of Lille2 collaborators