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.

Understanding consumers repurchase in the context of online shopping: an empirical study

L'utilisation de l'Internet comme medium pour faire ses courses et achats a vu une croissance exponentielle. Cependant, 99% des nouveaux business en ligne échouent. La plupart des acheteurs en ligne ne reviennent pas pour un ré-achat et 60% abandonnent leur chariot avant de conclure l'achat. En effet, après le premier achat, la rétention du consommateur en ligne devient critique au succès du vendeur de commerce électronique. Retenir des consommateurs peut sauver des coûts, accroître les profits, et permet de gagner un avantage compétitif.Les recherches précédentes ont identifié la loyauté comme étant le facteur le plus important dans la rétention du consommateur, et l'engagement ("commitment") comme étant un des facteurs les plus importants en marketing relationnel, offrant une réflexion sur la loyauté. Pourtant, nous n'avons pu trouver d'étude en commerce électronique examinant l'impact de la loyauté en ligne et de l'engagement en ligne ("online commitment") sur le ré-achat en ligne. Un des avantages de l'achat en ligne c'est la capacité à chercher le meilleur prix avec un clic. Pourtant, nous n'avons pu trouver de recherche empirique en commerce électronique qui examinait l'impact de la perception post-achat du prix sur le ré-achat en ligne.L'objectif de cette recherche est de développer un modèle théorique visant à comprendre le ré-achat en ligne, ou la continuité d'achat ("purchase continuance") du même magasin en ligne.Notre modèle de recherche a été testé dans un contexte de commerce électronique réel, sur un échantillon total de 1,866 vrais acheteurs d'un même magasin en ligne. L'étude est centrée sur le ré-achat. Par conséquent, les répondants sélectionnés aléatoirement devaient avoir acheté au moins une fois de ce magasin en ligne avant le début de l'enquête. Cinq mois plus tard, nous avons suivi les répondants pour voir s'ils étaient effectivement revenus pour un ré-achat.Notre analyse démontre que l'intention de ré-achat en ligne n'a pas d'impact significatif sur le ré-achat en ligne. La perception post-achat du prix en ligne ("post-purchase Price perception") et l'engagement normatif en ligne ("Normative Commitment") n'ont pas d'impact significatif sur l'intention de ré-achat en ligne. L'engagement affectif en ligne ("Affective Commitment"), l'attitude loyale en ligne ("Attitudinal Loyalty"), le comportement loyal en ligne ("Behavioral Loyalty"), l'engagement calculé en ligne ("Calculative Commitment") ont un impact positif sur l'intention de ré-achat en ligne. De plus, l'attitude loyale en ligne a un effet de médiation partielle entre l'engagement affectif en ligne et l'intention de ré-achat en ligne. Le comportement loyal en ligne a un effet de mediation partielle entre l'attitude loyale en ligne et l'intention de ré-achat en ligne.Nous avons réalisé deux analyses complémentaires : 1) Sur un échantillon de premiers acheteurs, nous trouvons que la perception post-achat du prix en ligne a un impact positif sur l'intention de ré-achat en ligne. 2) Nous avons divisé l'échantillon de l'étude principale entre des acheteurs répétitifs Suisse-Romands et Suisse-Allemands. Les résultats démontrent que les Suisse-Romands montrent plus d'émotions durant l'achat en ligne que les Suisse-Allemands. Nos résultats contribuent à la recherche académique mais aussi aux praticiens de l'industrie e-commerce.AbstractThe use of the Internet as a shopping and purchasing medium has seen exceptional growth. However, 99% of new online businesses fail. Most online buyers do not comeback for a repurchase, and 60% abandon their shopping cart before checkout. Indeed, after the first purchase, online consumer retention becomes critical to the success of the e-commerce vendor. Retaining existing customers can save costs, increase profits, and is a means of gaining competitive advantage.Past research identified loyalty as the most important factor in achieving customer retention, and commitment as one of the most important factors in relationship marketing, providing a good description of what type of thinking leads to loyalty. Yet, we could not find an e-commerce study investing the impact of both online loyalty and online commitment on online repurchase. One of the advantages of online shopping is the ability of browsing for the best price with one click. Yet, we could not find an e- commerce empirical research investigating the impact of post-purchase price perception on online repurchase.The objective of this research is to develop a theoretical model aimed at understanding online repurchase, or purchase continuance from the same online store.Our model was tested in a real e-commerce context with an overall sample of 1, 866 real online buyers from the same online store.The study focuses on repurchase. Therefore, randomly selected respondents had purchased from the online store at least once prior to the survey. Five months later, we tracked respondents to see if they actually came back for a repurchase.Our findings show that online Intention to repurchase has a non-significant impact on online Repurchase. Online post-purchase Price perception and online Normative Commitment have a non-significant impact on online Intention to repurchase, whereas online Affective Commitment, online Attitudinal Loyalty, online Behavioral Loyalty, and online Calculative Commitment have a positive impact on online Intention to repurchase. Furthermore, online Attitudinal Loyalty partially mediates between online Affective Commitment and online Intention to repurchase, and online Behavioral Loyalty partially mediates between online Attitudinal Loyalty and online Intention to repurchase.We conducted two follow up analyses: 1) On a sample of first time buyers, we find that online post-purchase Price perception has a positive impact on Intention. 2) We divided the main study's sample into Swiss-French and Swiss-German repeated buyers. Results show that Swiss-French show more emotions when shopping online than Swiss- Germans. Our findings contribute to academic research but also to practice.

www.fevertravel.ch : an online study prototype to evaluate the safety and feasibility of computerized guidelines for fever in returning travellers and migrants

Résumé Suite à la publication des recommandations de pratiques cliniques sur la prise en soins de patients fébriles au retour des tropiques, nous avons développé un site internet de consultation www.fevertravel.ch qui comprend un arbre décisionnel et des caractéristiques diagnostiques spécifiques fournissant une assistance diagnostique aux médecins de premier recours. Nous avons ensuite intégré une composante de recherche afin d'évaluer l'implémentation de ces recommandations de pratiques cliniques informatisées. De plus, le site est capable d'enregistrer : (1) le chemin parcouru par le médecin au travers de l'arbre décisionnel, (2) les tests diagnostics effectués, (3) les diagnostics initial et final ainsi que les devenirs des patients et (4) les raisons de non-adhérence lorsque les médecins divergent de l'attitude proposée. Nous croyons que la technologie internat est un moyen puissant pour atteindre des médecins provenant de différents horizons dans leur propre environnement et qu'il pourrait se montrer être un outil de recherche efficace pour disséminer les recommandations de pratiques cliniques et évaluer leur justesse-adéquation. Dans cet article, nous décrivons le design, le contenu, l'architecture et l'implémentation du système de ce prototype d'étude interactive qui vise à intégrer une recherche opérationnelle en médecine de premier recours.

Do resources, justice administration practices and federalism have an impact on registered and sentenced crime prevalence ?

This contribution, based on a statistical approach, undertakes to link data on resources (personnel and financial means) and the working of the administration of penal justice (prosecution, sentencing) taking into account the nationality of those prosecuted. In order to be able to distinguish prosecution and sentencing practices of judicial authorities and possible processes of discrimination, diverse sources have been used such as data from court administrations, public finances and police forces, collected by the Swiss Federal Statistical Office and the Swiss Federal administration of finances. The authors discuss discrimination in prosecution and sentencing between Swiss residents and foreigners taking into account localization and resources regarding personnel and public finances.

Comparison of HIV-infected patients' characteristics, healthcare resources use and cost between native and migrant patients

OBJECTIVES: To assess whether patients' characteristics and healthcare resources consumption and costs were different between native and migrant populations in Switzerland. METHODS: All adult patients followed-up in the Swiss HIV-cohort study in our institution during 2000-2003 were considered. Patients' characteristics were retrieved from the cohort database. Hospital and outpatient resource use were extracted from individual charts and valued with 2002 tariffs. RESULTS: The 66 migrants were younger (29 +/- 8 years versus 37 +/- 11, p < 0.001), less often of male gender (38 % versus 70 %, p < 0.001), predominantly infected via heterosexual contact (87 % versus 52 %, p < 0.01), with lower mean CD4 level at enrollment (326 +/- 235 versus 437 +/- 305, p = 0.002) than their 200 native counterparts. Migrants had fewer hospitalizations, more frequent outpatient visits, laboratory tests, and lower total cost of care per year of follow-up (<euro> 2'215 +/- 4'206 versus 4'155 +/- 12'304, p = 0.037). Resource use and costs were significantly higher in people with < 200 CD4 cell counts in both groups. CONCLUSIONS: Migrant population had more advanced disease, more outpatient visits but less hospitalizations, resulting in lower costs of care when compared with native population.

Evaluation of an online, case-based interactive approach to teaching pathophysiology

AIM: The aim of this study was to evaluate a new pedagogical approach in teaching fluid, electrolyte and acid-base pathophysiology in undergraduate students. METHODS: This approach comprises traditional lectures, the study of clinical cases on the web and a final interactive discussion of these cases in the classroom. When on the web, the students are asked to select laboratory tests that seem most appropriate to understand the pathophysiological condition underlying the clinical case. The percentage of students having chosen a given test is made available to the teacher who uses it in an interactive session to stimulate discussion with the whole class of students. The same teacher used the same case studies during 2 consecutive years during the third year of the curriculum. RESULTS: The majority of students answered the questions on the web as requested and evaluated positively their experience with this form of teaching and learning. CONCLUSIONS: Complementing traditional lectures with online case-based studies and interactive group discussions represents, therefore, a simple means to promote the learning and the understanding of complex pathophysiological mechanisms. This simple problem-based approach to teaching and learning may be implemented to cover all fields of medicine.

Profile of European radiotherapy departments contributing to the EORTC Radiation Oncology Group (ROG) in the 21st century.

PURPOSE: Since 1982, the Radiation Oncology Group of the EORTC (EORTC ROG) has pursued an extensive Quality Assurance (QA) program involving all centres actively participating in its clinical research. The first step is the evaluation of the structure and of the human, technical and organisational resources of the centres, to assess their ability to comply with the current requirements for high-tech radiotherapy (RT). MATERIALS AND METHODS: A facility questionnaire (FQ) was developed in 1989 and adapted over the years to match the evolution of RT techniques. We report on the contents of the current FQ that was completed online by 98 active EORTC ROG member institutions from 19 countries, between December 2005 and October 2007. RESULTS: Similar to the data collected previously, large variations in equipment, staffing and workload between centres remain. Currently only 15 centres still use a Cobalt unit. All centres perform 3D Conformal RT, 79% of them can perform IMRT and 54% are able to deliver stereotactic RT. An external reference dosimetry audit (ERDA) was performed in 88% of the centres for photons and in 73% for electrons, but it was recent (<2 years) in only 74% and 60%, respectively. CONCLUSION: The use of the FQ helps maintain the minimum quality requirements within the EORTC ROG network: recommendations are made on the basis of the analysis of its results. The present analysis shows that modern RT techniques are widely implemented in the clinic but also that ERDA should be performed more frequently. Repeated assessment using the FQ is warranted to document the future evolution of the EORTC ROG institutions.

Use of mobile devices to answer online surveys: implications for research.

BACKGROUND: There is a growing use of mobile devices to access the Internet. We examined whether participants who used a mobile device to access a brief online survey were quicker to respond to the survey but also, less likely to complete it than participants using a traditional web browser. FINDINGS: Using data from a recently completed online intervention trial, we found that participants using mobile devices were quicker to access the survey but less likely to complete it compared to participants using a traditional web browser. More concerning, mobile device users were also less likely to respond to a request to complete a six week follow-up survey compared to those using traditional web browsers. CONCLUSIONS: With roughly a third of participants using mobile devices to answer an online survey in this study, the impact of mobile device usage on survey completion rates is a concern. TRIAL REGISTRATION: ClinicalTrials.gov: NCT01521078.