Modèle et application pour la génération automatique de parcours d'apprentissage personnalisés à partir de sources de confiance

Résumé Lors d'une recherche d'information, l'apprenant est très souvent confronté à des problèmes de guidage et de personnalisation. Ceux-ci sont d'autant plus importants que la recherche se fait dans un environnement ouvert tel que le Web. En effet, dans ce cas, il n'y a actuellement pas de contrôle de pertinence sur les ressources proposées pas plus que sur l'adéquation réelle aux besoins spécifiques de l'apprenant. A travers l'étude de l'état de l'art, nous avons constaté l'absence d'un modèle de référence qui traite des problématiques liées (i) d'une part aux ressources d'apprentissage notamment à l'hétérogénéité de la structure et de la description et à la protection en terme de droits d'auteur et (ii) d'autre part à l'apprenant en tant qu'utilisateur notamment l'acquisition des éléments le caractérisant et la stratégie d'adaptation à lui offrir. Notre objectif est de proposer un système adaptatif à base de ressources d'apprentissage issues d'un environnement à ouverture contrôlée. Celui-ci permet de générer automatiquement sans l'intervention d'un expert pédagogue un parcours d'apprentissage personnalisé à partir de ressources rendues disponibles par le biais de sources de confiance. L'originalité de notre travail réside dans la proposition d'un modèle de référence dit de Lausanne qui est basé sur ce que nous considérons comme étant les meilleures pratiques des communautés : (i) du Web en terme de moyens d'ouverture, (ii) de l'hypermédia adaptatif en terme de stratégie d'adaptation et (iii) de l'apprentissage à distance en terme de manipulation des ressources d'apprentissage. Dans notre modèle, la génération des parcours personnalisés se fait sur la base (i) de ressources d'apprentissage indexées et dont le degré de granularité en favorise le partage et la réutilisation. Les sources de confiance utilisées en garantissent l'utilité et la qualité. (ii) de caractéristiques de l'utilisateur, compatibles avec les standards existants, permettant le passage de l'apprenant d'un environnement à un autre. (iii) d'une adaptation à la fois individuelle et sociale. Pour cela, le modèle de Lausanne propose : (i) d'utiliser ISO/MLR (Metadata for Learning Resources) comme formalisme de description. (ii) de décrire le modèle d'utilisateur avec XUN1 (eXtended User Model), notre proposition d'un modèle compatible avec les standards IEEE/PAPI et IMS/LIP. (iii) d'adapter l'algorithme des fourmis au contexte de l'apprentissage à distance afin de générer des parcours personnalisés. La dimension individuelle est aussi prise en compte par la mise en correspondance de MLR et de XUM. Pour valider notre modèle, nous avons développé une application et testé plusieurs scenarii mettant en action des utilisateurs différents à des moments différents. Nous avons ensuite procédé à des comparaisons entre ce que retourne le système et ce que suggère l'expert. Les résultats s'étant avérés satisfaisants dans la mesure où à chaque fois le système retourne un parcours semblable à celui qu'aurait proposé l'expert, nous sommes confortées dans notre approche.

Machine learning algorithms for spatial data. Case studies: Environmental pollution, natural hazards, renewable resources

This paper presents general problems and approaches for the spatial data analysis using machine learning algorithms. Machine learning is a very powerful approach to adaptive data analysis, modelling and visualisation. The key feature of the machine learning algorithms is that they learn from empirical data and can be used in cases when the modelled environmental phenomena are hidden, nonlinear, noisy and highly variable in space and in time. Most of the machines learning algorithms are universal and adaptive modelling tools developed to solve basic problems of learning from data: classification/pattern recognition, regression/mapping and probability density modelling. In the present report some of the widely used machine learning algorithms, namely artificial neural networks (ANN) of different architectures and Support Vector Machines (SVM), are adapted to the problems of the analysis and modelling of geo-spatial data. Machine learning algorithms have an important advantage over traditional models of spatial statistics when problems are considered in a high dimensional geo-feature spaces, when the dimension of space exceeds 5. Such features are usually generated, for example, from digital elevation models, remote sensing images, etc. An important extension of models concerns considering of real space constrains like geomorphology, networks, and other natural structures. Recent developments in semi-supervised learning can improve modelling of environmental phenomena taking into account on geo-manifolds. An important part of the study deals with the analysis of relevant variables and models' inputs. This problem is approached by using different feature selection/feature extraction nonlinear tools. To demonstrate the application of machine learning algorithms several interesting case studies are considered: digital soil mapping using SVM, automatic mapping of soil and water system pollution using ANN; natural hazards risk analysis (avalanches, landslides), assessments of renewable resources (wind fields) with SVM and ANN models, etc. The dimensionality of spaces considered varies from 2 to more than 30. Figures 1, 2, 3 demonstrate some results of the studies and their outputs. Finally, the results of environmental mapping are discussed and compared with traditional models of geostatistics.

A Survey of Active Learning Algorithms for Supervised Remote Sensing Image Classification

Defining an efficient training set is one of the most delicate phases for the success of remote sensing image classification routines. The complexity of the problem, the limited temporal and financial resources, as well as the high intraclass variance can make an algorithm fail if it is trained with a suboptimal dataset. Active learning aims at building efficient training sets by iteratively improving the model performance through sampling. A user-defined heuristic ranks the unlabeled pixels according to a function of the uncertainty of their class membership and then the user is asked to provide labels for the most uncertain pixels. This paper reviews and tests the main families of active learning algorithms: committee, large margin, and posterior probability-based. For each of them, the most recent advances in the remote sensing community are discussed and some heuristics are detailed and tested. Several challenging remote sensing scenarios are considered, including very high spatial resolution and hyperspectral image classification. Finally, guidelines for choosing the good architecture are provided for new and/or unexperienced user.

On optimal learning schedules and the marginal value of cumulative cultural evolution.

The age-dependent choice between expressing individual learning (IL) or social learning (SL) affects cumulative cultural evolution. A learning schedule in which SL precedes IL is supportive of cumulative culture because the amount of nongenetically encoded adaptive information acquired by previous generations can be absorbed by an individual and augmented. Devoting time and energy to learning, however, reduces the resources available for other life-history components. Learning schedules and life history thus coevolve. Here, we analyze a model where individuals may have up to three distinct life stages: "infants" using IL or oblique SL, "juveniles" implementing IL or horizontal SL, and adults obtaining material resources with learned information. We study the dynamic allocation of IL and SL within life stages and how this coevolves with the length of the learning stages. Although no learning may be evolutionary stable, we find conditions where cumulative cultural evolution can be selected for. In that case, the evolutionary stable learning schedule causes individuals to use oblique SL during infancy and a mixture between IL and horizontal SL when juvenile. We also find that the selected pattern of oblique SL increases the amount of information in the population, but horizontal SL does not do so.

Multitask learning of environmental spatial data

The present research deals with an application of artificial neural networks for multitask learning from spatial environmental data. The real case study (sediments contamination of Geneva Lake) consists of 8 pollutants. There are different relationships between these variables, from linear correlations to strong nonlinear dependencies. The main idea is to construct a subsets of pollutants which can be efficiently modeled together within the multitask framework. The proposed two-step approach is based on: 1) the criterion of nonlinear predictability of each variable ?k? by analyzing all possible models composed from the rest of the variables by using a General Regression Neural Network (GRNN) as a model; 2) a multitask learning of the best model using multilayer perceptron and spatial predictions. The results of the study are analyzed using both machine learning and geostatistical tools.

Se former à l'entretien motivationnel pour aider ses patients à changer: Les étudiants en médecine le confirment [Learning motivational interviewing to help patients change their health-related behaviors: medical students confirm it].

At the Lausanne University, 5th year medical students were trained in Motivational interviewing (MI). Eight hours of training improved their competence in the use of this approach. This experience supports the implementation of MI training in medical schools. Motivational interviewing allows the health professional to actively involve the patient in this behavior change process (drinking, smoking, diet, exercise, medication adherence, etc.), by encouraging reflection and reinforcing personal motivation and resources.

A two-week workshop to promote cardiovascular disease prevention programs in countries with limited resources.

Training is a crucial tool for building the capacity necessary for prevention and control of cardiovascular diseases (CVDs) in developing countries. This paper summarizes some features of a 2-week workshop aimed at enabling local health professionals to initiate a comprehensive CVD prevention and control program in a context of limited resources. The workshops have been organized in the regions where CVD prevention programs are being contemplated, in cooperation with health authorities of the concerned regions. The workshop's content includes a broad variety of issues related to CVD prevention and control, and to program development. Strong emphasis is placed on "learning by doing," and groups of 5-6 participants conduct a small-scale epidemiological study during the first week; during the second week, they draft a virtual program of CVD prevention and control adapted to the local situation. This practice-oriented workshop focuses on building expertise among anticipated key players, strengthening networks among relevant health professionals, and advocating the urgent need to tackle the emerging CVD epidemic in developing countries.

Learning from failure - rationale and design for a study about discontinuation of randomized trials (DISCO study).

BACKGROUND: Randomized controlled trials (RCTs) may be discontinued because of apparent harm, benefit, or futility. Other RCTs are discontinued early because of insufficient recruitment. Trial discontinuation has ethical implications, because participants consent on the premise of contributing to new medical knowledge, Research Ethics Committees (RECs) spend considerable effort reviewing study protocols, and limited resources for conducting research are wasted. Currently, little is known regarding the frequency and characteristics of discontinued RCTs. METHODS/DESIGN: Our aims are, first, to determine the prevalence of RCT discontinuation for specific reasons; second, to determine whether the risk of RCT discontinuation for specific reasons differs between investigator- and industry-initiated RCTs; third, to identify risk factors for RCT discontinuation due to insufficient recruitment; fourth, to determine at what stage RCTs are discontinued; and fifth, to examine the publication history of discontinued RCTs.We are currently assembling a multicenter cohort of RCTs based on protocols approved between 2000 and 2002/3 by 6 RECs in Switzerland, Germany, and Canada. We are extracting data on RCT characteristics and planned recruitment for all included protocols. Completion and publication status is determined using information from correspondence between investigators and RECs, publications identified through literature searches, or by contacting the investigators. We will use multivariable regression models to identify risk factors for trial discontinuation due to insufficient recruitment. We aim to include over 1000 RCTs of which an anticipated 150 will have been discontinued due to insufficient recruitment. DISCUSSION: Our study will provide insights into the prevalence and characteristics of RCTs that were discontinued. Effective recruitment strategies and the anticipation of problems are key issues in the planning and evaluation of trials by investigators, Clinical Trial Units, RECs and funding agencies. Identification and modification of barriers to successful study completion at an early stage could help to reduce the risk of trial discontinuation, save limited resources, and enable RCTs to better meet their ethical requirements.