Machine learning for geospatial data : algorithms, software tools and case studies

Résumé Cette thèse est consacrée à l'analyse, la modélisation et la visualisation de données environnementales à référence spatiale à l'aide d'algorithmes d'apprentissage automatique (Machine Learning). L'apprentissage automatique peut être considéré au sens large comme une sous-catégorie de l'intelligence artificielle qui concerne particulièrement le développement de techniques et d'algorithmes permettant à une machine d'apprendre à partir de données. Dans cette thèse, les algorithmes d'apprentissage automatique sont adaptés pour être appliqués à des données environnementales et à la prédiction spatiale. Pourquoi l'apprentissage automatique ? Parce que la majorité des algorithmes d'apprentissage automatiques sont universels, adaptatifs, non-linéaires, robustes et efficaces pour la modélisation. Ils peuvent résoudre des problèmes de classification, de régression et de modélisation de densité de probabilités dans des espaces à haute dimension, composés de variables informatives spatialisées (« géo-features ») en plus des coordonnées géographiques. De plus, ils sont idéaux pour être implémentés en tant qu'outils d'aide à la décision pour des questions environnementales allant de la reconnaissance de pattern à la modélisation et la prédiction en passant par la cartographie automatique. Leur efficacité est comparable au modèles géostatistiques dans l'espace des coordonnées géographiques, mais ils sont indispensables pour des données à hautes dimensions incluant des géo-features. Les algorithmes d'apprentissage automatique les plus importants et les plus populaires sont présentés théoriquement et implémentés sous forme de logiciels pour les sciences environnementales. Les principaux algorithmes décrits sont le Perceptron multicouches (MultiLayer Perceptron, MLP) - l'algorithme le plus connu dans l'intelligence artificielle, le réseau de neurones de régression généralisée (General Regression Neural Networks, GRNN), le réseau de neurones probabiliste (Probabilistic Neural Networks, PNN), les cartes auto-organisées (SelfOrganized Maps, SOM), les modèles à mixture Gaussiennes (Gaussian Mixture Models, GMM), les réseaux à fonctions de base radiales (Radial Basis Functions Networks, RBF) et les réseaux à mixture de densité (Mixture Density Networks, MDN). Cette gamme d'algorithmes permet de couvrir des tâches variées telle que la classification, la régression ou l'estimation de densité de probabilité. L'analyse exploratoire des données (Exploratory Data Analysis, EDA) est le premier pas de toute analyse de données. Dans cette thèse les concepts d'analyse exploratoire de données spatiales (Exploratory Spatial Data Analysis, ESDA) sont traités selon l'approche traditionnelle de la géostatistique avec la variographie expérimentale et selon les principes de l'apprentissage automatique. La variographie expérimentale, qui étudie les relations entre pairs de points, est un outil de base pour l'analyse géostatistique de corrélations spatiales anisotropiques qui permet de détecter la présence de patterns spatiaux descriptible par une statistique. L'approche de l'apprentissage automatique pour l'ESDA est présentée à travers l'application de la méthode des k plus proches voisins qui est très simple et possède d'excellentes qualités d'interprétation et de visualisation. Une part importante de la thèse traite de sujets d'actualité comme la cartographie automatique de données spatiales. Le réseau de neurones de régression généralisée est proposé pour résoudre cette tâche efficacement. Les performances du GRNN sont démontrées par des données de Comparaison d'Interpolation Spatiale (SIC) de 2004 pour lesquelles le GRNN bat significativement toutes les autres méthodes, particulièrement lors de situations d'urgence. La thèse est composée de quatre chapitres : théorie, applications, outils logiciels et des exemples guidés. Une partie importante du travail consiste en une collection de logiciels : Machine Learning Office. Cette collection de logiciels a été développée durant les 15 dernières années et a été utilisée pour l'enseignement de nombreux cours, dont des workshops internationaux en Chine, France, Italie, Irlande et Suisse ainsi que dans des projets de recherche fondamentaux et appliqués. Les cas d'études considérés couvrent un vaste spectre de problèmes géoenvironnementaux réels à basse et haute dimensionnalité, tels que la pollution de l'air, du sol et de l'eau par des produits radioactifs et des métaux lourds, la classification de types de sols et d'unités hydrogéologiques, la cartographie des incertitudes pour l'aide à la décision et l'estimation de risques naturels (glissements de terrain, avalanches). Des outils complémentaires pour l'analyse exploratoire des données et la visualisation ont également été développés en prenant soin de créer une interface conviviale et facile à l'utilisation. Machine Learning for geospatial data: algorithms, software tools and case studies Abstract The thesis is devoted to the analysis, modeling and visualisation of spatial environmental data using machine learning algorithms. In a broad sense machine learning can be considered as a subfield of artificial intelligence. It mainly concerns with the development of techniques and algorithms that allow computers to learn from data. In this thesis machine learning algorithms are adapted to learn from spatial environmental data and to make spatial predictions. Why machine learning? In few words most of machine learning algorithms are universal, adaptive, nonlinear, robust and efficient modeling tools. They can find solutions for the classification, regression, and probability density modeling problems in high-dimensional geo-feature spaces, composed of geographical space and additional relevant spatially referenced features. They are well-suited to be implemented as predictive engines in decision support systems, for the purposes of environmental data mining including pattern recognition, modeling and predictions as well as automatic data mapping. They have competitive efficiency to the geostatistical models in low dimensional geographical spaces but are indispensable in high-dimensional geo-feature spaces. The most important and popular machine learning algorithms and models interesting for geo- and environmental sciences are presented in details: from theoretical description of the concepts to the software implementation. The main algorithms and models considered are the following: multi-layer perceptron (a workhorse of machine learning), general regression neural networks, probabilistic neural networks, self-organising (Kohonen) maps, Gaussian mixture models, radial basis functions networks, mixture density networks. This set of models covers machine learning tasks such as classification, regression, and density estimation. Exploratory data analysis (EDA) is initial and very important part of data analysis. In this thesis the concepts of exploratory spatial data analysis (ESDA) is considered using both traditional geostatistical approach such as_experimental variography and machine learning. Experimental variography is a basic tool for geostatistical analysis of anisotropic spatial correlations which helps to understand the presence of spatial patterns, at least described by two-point statistics. A machine learning approach for ESDA is presented by applying the k-nearest neighbors (k-NN) method which is simple and has very good interpretation and visualization properties. Important part of the thesis deals with a hot topic of nowadays, namely, an automatic mapping of geospatial data. General regression neural networks (GRNN) is proposed as efficient model to solve this task. Performance of the GRNN model is demonstrated on Spatial Interpolation Comparison (SIC) 2004 data where GRNN model significantly outperformed all other approaches, especially in case of emergency conditions. The thesis consists of four chapters and has the following structure: theory, applications, software tools, and how-to-do-it examples. An important part of the work is a collection of software tools - Machine Learning Office. Machine Learning Office tools were developed during last 15 years and was used both for many teaching courses, including international workshops in China, France, Italy, Ireland, Switzerland and for realizing fundamental and applied research projects. Case studies considered cover wide spectrum of the real-life low and high-dimensional geo- and environmental problems, such as air, soil and water pollution by radionuclides and heavy metals, soil types and hydro-geological units classification, decision-oriented mapping with uncertainties, natural hazards (landslides, avalanches) assessments and susceptibility mapping. Complementary tools useful for the exploratory data analysis and visualisation were developed as well. The software is user friendly and easy to use.

Schemata as a framework for full body and physical design interfaces

Digital art interfaces presents cognitiveparadigms that deals with the recognition of the symbols and representations through interaction.What is presented in this paper is anapproximation of the bodily experience in that particular scenario and a new proposal which has the aim to contribute more ideas and criteria in the analysis of the learning process of aparticipant discovering an interactive space or interface. For that I propose a first new approach where metaphorically I tried to extrapolate the stages of the psychology of development stated byJean Piaget in the interface design domain.

Milk flow, teat morphology and subclinical mastitis prevalence in Gir cows

The aim of this work was to evaluate the association between milk flow, teat morphological measurements and subclinical mastitis prevalence in Gir cows. Eighty cows in the 2nd and 3rd lactations, with 90 to 200 days of lactation, were divided according to milk flow during milking into fast or slow groups. Teat morphometry was assessed by ultrasound scanning of the right anterior teat and external measurements. Milk samples were collected for somatic cells count (SCC) and microbiological culture. The effect of milk flow during milking was evaluated by analysis of variance of milk yield, SCC, morphometry and external measurements. The association of morphometry and external measurements of the teats with the SCC and microorganisms found in milk were analysed. Milk flow was significantly correlated to milk production. Gir cows with slower milk flow had longer teat canal and greater milk yield, in comparison to cows with fast milk flow. Teat-end to floor distance influenced SCC of Gir cows. Prevalence of subclinical mastitis and the type of mastitis-causing pathogens were not affected by milk flow during milking

In vitro conservation of Piper aduncum and Piper hispidinervum under slow-growth conditions

The objective of this work was to evaluate in vitro storage of Piper aduncum and P. hispidinervum under slow-growth conditions. Shoots were stored at low temperatures (10, 20 and 25°C), and the culture medium was supplemented with osmotic agents (sucrose and mannitol - at 1, 2 and 3%) and abiscisic acid - ABA (0, 0.5, 1.0, 2.0 and 3.0 mg L-1). After six-months of storage, shoots were evaluated for survival and regrowth. Low temperature at 20ºC was effective for the in vitro conservation of P. aduncum and P. hispidinervum shoots. In vitro cultures maintained at 20ºC on MS medium showed 100% survival with slow-growth shoots. The presence of mannitol or ABA, in the culture medium, negatively affected shoot growth, which is evidenced by the low rate of recovered shoots.

Hes1 is a critical but context-dependent mediator of canonical Notch signaling in lymphocyte development and transformation.

Although canonical Notch signaling regulates multiple hematopoietic lineage decisions including T cell and marginal zone B cell fate specification, the downstream molecular mediators of Notch function are largely unknown. We showed here that conditional inactivation of Hes1, a well-characterized Notch target gene, in adult murine bone marrow (BM) cells severely impaired T cell development without affecting other Notch-dependent hematopoietic lineages such as marginal zone B cells. Competitive mixed BM chimeras, intrathymic transfer experiments, and in vitro culture of BM progenitors on Delta-like-expressing stromal cells further demonstrated that Hes1 is required for T cell lineage commitment, but dispensable for Notch-dependent thymocyte maturation through and beyond the beta selection checkpoint. Furthermore, our data strongly suggest that Hes1 is essential for the development and maintenance of Notch-induced T cell acute lymphoblastic leukemia. Collectively, our studies identify Hes1 as a critical but context-dependent mediator of canonical Notch signaling in the hematopoietic system.

Usability of Virtual Museums and the Diffusion of Cultural Heritage

Les prioritats per als museus canvien. La missió de la nova museologia és convertir els museus en llocs per a gaudir i aprendre, cosa que fa que hagin de dur a terme una gestió financera molt semblant a la d'una empresa social que competeixi en el sector del lleure. Amb el pas del temps, els museus han d'establir i aplicar els criteris necessaris per a la supervivència, aplanant el terreny perquè altres institucions públiques siguin més obertes en els seus esforços per comunicar i difondre el seu patrimoni. Ja podem començar a parlar d'algunes conclusions comunament acceptades sobre el comportament dels visitants, que són necessàries per a planificar exposicions futures que vegin l'aprenentatge com un procés constructiu, les col·leccions com a objectes amb significat i les mateixes exposicions com a mitjans de comunicació que haurien de transformar la manera de pensar de l'espectador i que estan al servei del mateix missatge. Sembla que internet representa un mitjà efectiu per a assolir aquests objectius, ja que és capaç (a) d'adaptar-se als interessos i les característiques intel·lectuals d'un públic divers; (b) de redescobrir els significats dels objectes i adquirir un reconeixement sociocultural del seu valor per mitjà del seu potencial interactiu, i (c) de fer ús d'elements atractius i estimulants perquè tothom en gaudeixi. Per a aquest propòsit, és bàsic fer-nos les preguntes següents: quins criteris ha de seguir un museu virtual per a optimar la difusió del seu patrimoni?; quins elements estimulen els usuaris a quedar-se en una pàgina web i fer visites virtuals que els siguin satisfactòries?; quin paper té la usabilitat de l'aplicació en tot això?

The relationships of organizational cultures, trust and innovativeness

Sekä organisaatiokulttuuria, luottamusta että innovatiivisuutta on tutkittu paljon, mutta toistaiseksi nämä käsitteet yhdistävää kokonaisvaltaista tutkimusta ei juuri ole tehty tai ainakaan raportoitu tieteellisissä julkaisuissa. Tätä tutkimus käsitteli organisaatiokulttuurin, luottamuksen ja innovatiivisuuden suhteita erilaisissa organisaatiokulttuureissa. Tutkimuksen tavoitteena oli tutkia organisaatiokulttuurin vaikutusta luottamukseen (sekä kompetenssiin, hyväntahtoisuuteen että rehellisyyteen perustuvaan lateraaliin, vertikaaliin ja institutionaaliseen luottamukseen), innovaatioilmastoon ja innovaatiotoiminnan tuloksellisuuteen. Organisaatiokulttuurin, luottamuksen ja innovatiivisuuden yhteyttä tarkasteltiin neljässä erityyppisessä organisaatiokulttuurissa (klaani-, adhokratia-, hierarkia- ja markkinakulttuurit), jotka pohjautuvat kilpailevien arvojen malliin. Tutkimuksen empiirinen osa toteutettiin posti ja Internet -pohjaisena kyselytutkimuksena 40 organisaatioyksikössä tilastollisen analyysin menetelmin. Yleisellä tasolla työssä saatiin selville, että luottamuksen ja innovatiivisuuden tasot vaihtelevat erityyppisissä organisaatio-kulttuureissa. Tarkemmin sanottuna klaani- ja adhokratiakulttuureissa luottamus ja innovatiivisuus olivat korkeita, ja näillä kulttuureilla oli myös positiivinen vaikutus innovaatiotoiminnan tuloksellisuuteen. Erityisesti institutionaalisen luottamuksen ja innovaatiotuen merkitykset olivat tärkeitä, sillä ne toimivat mediaattoreina organisaatiokulttuurin ja innovatiivisuuden välisessä suhteessa. Luottamuksella ja innovatiivisuudelle ei sitä vastoin ollut vaikutusta hierarkia- ja markkinakulttuureissa, tai vaikutus oli negatiivinen. Tässä työssä osoitettiin myös aiemmin hyvin vähän tutkitun institutionaalisen organisatorisen luottamuksen merkitys organisaatioiden innovatiivisuudessa.

Reconstructing native American migrations from whole-genome and whole-exome data.

There is great scientific and popular interest in understanding the genetic history of populations in the Americas. We wish to understand when different regions of the continent were inhabited, where settlers came from, and how current inhabitants relate genetically to earlier populations. Recent studies unraveled parts of the genetic history of the continent using genotyping arrays and uniparental markers. The 1000 Genomes Project provides a unique opportunity for improving our understanding of population genetic history by providing over a hundred sequenced low coverage genomes and exomes from Colombian (CLM), Mexican-American (MXL), and Puerto Rican (PUR) populations. Here, we explore the genomic contributions of African, European, and especially Native American ancestry to these populations. Estimated Native American ancestry is 48% in MXL, 25% in CLM, and 13% in PUR. Native American ancestry in PUR is most closely related to populations surrounding the Orinoco River basin, confirming the Southern American ancestry of the Taíno people of the Caribbean. We present new methods to estimate the allele frequencies in the Native American fraction of the populations, and model their distribution using a demographic model for three ancestral Native American populations. These ancestral populations likely split in close succession: the most likely scenario, based on a peopling of the Americas 16 thousand years ago (kya), supports that the MXL Ancestors split 12.2kya, with a subsequent split of the ancestors to CLM and PUR 11.7kya. The model also features effective populations of 62,000 in Mexico, 8,700 in Colombia, and 1,900 in Puerto Rico. Modeling Identity-by-descent (IBD) and ancestry tract length, we show that post-contact populations also differ markedly in their effective sizes and migration patterns, with Puerto Rico showing the smallest effective size and the earlier migration from Europe. Finally, we compare IBD and ancestry assignments to find evidence for relatedness among European founders to the three populations.

Género y cultura popular

En uno de los capítulos de Los Simpson, la pequeña Lisa compra, emocionada, la última versión de Stacy Malibú (el equivalente a nuestra Barbie). La novedad consiste en que después de 50 años de existencia, la muñeca habla. Pero para consternación de Lisa, el repertorio de frases es de lo más indignante: "Me encantaría que en la escuela enseñaran a ir de compras", "¡Vamos a hornear unas galletas para los chicos!" o "No me preguntes: sólo soy una chica (risita vacua)". Obviamente, la comprometida y concienciada Lisa no puede quedarse de brazos cruzados y pide a su madre que la lleve a la fábrica de Stacy Malibú para presentar sus quejas. La conversación entre ambas es enormemente reveladora: aunque Marge apoya a su hija e insiste en que siempre defienda sus ideas, considera que quizás está yendo más allá de lo razonable en su denuncia de que toda una generación de niñas se comportará como Stacy Malibú y la tomará como modelo y, finalmente, concluye en tono conciliador: "Yo tuve una Stacy Malibú de pequeña y no me ha pasado nada.¡Vamos a olvidarnos de estos problemas con un gran bol de helado de fresa!". La respuesta de Lisa no deja lugar a dudas; poniendo en funcionamiento a la muñeca y situándola frente a su madre, oímos a Stacy diciendo: "¡Vamos a olvidarnos de estos problemas con un gran bol de helado de fresa!". Huelgan los comentarios.

Género y cultura popular

En uno de los capítulos de Los Simpson, la pequeña Lisa compra, emocionada, la última versión de Stacy Malibú (el equivalente a nuestra Barbie). La novedad consiste en que después de 50 años de existencia, la muñeca habla. Pero para consternación de Lisa, el repertorio de frases es de lo más indignante: "Me encantaría que en la escuela enseñaran a ir de compras", "¡Vamos a hornear unas galletas para los chicos!" o "No me preguntes: sólo soy una chica (risita vacua)". Obviamente, la comprometida y concienciada Lisa no puede quedarse de brazos cruzados y pide a su madre que la lleve a la fábrica de Stacy Malibú para presentar sus quejas. La conversación entre ambas es enormemente reveladora: aunque Marge apoya a su hija e insiste en que siempre defienda sus ideas, considera que quizás está yendo más allá de lo razonable en su denuncia de que toda una generación de niñas se comportará como Stacy Malibú y la tomará como modelo y, finalmente, concluye en tono conciliador: "Yo tuve una Stacy Malibú de pequeña y no me ha pasado nada.¡Vamos a olvidarnos de estos problemas con un gran bol de helado de fresa!". La respuesta de Lisa no deja lugar a dudas; poniendo en funcionamiento a la muñeca y situándola frente a su madre, oímos a Stacy diciendo: "¡Vamos a olvidarnos de estos problemas con un gran bol de helado de fresa!". Huelgan los comentarios.

Courts and Courtly Cultures in Early Modern Italy and Europe: Models and Languages

This volume deals with the forms of propaganda and self-representation, through words and images, during the rise of the 'civiltà delle corti' and through processes typical of the time, such as confrontation, adaptation, competition and rivalry. This period, which marked the passage of Italian and European culture from the Middle Ages to the Renaissance, is fundamental in the development of Modern Europe, and it lasted up to the XVIII century and beyond. At the heart of many matters debated here lies the relationship between culture and politics. The formation of a 'Lombard identity', central to the Sinergia project which was the frame of the whole research and its conferences, is closely linked to this broad general context. It places the so called 'questione milanese' - above the traditional hierarchies 'Toscana oriented' - at the centre of many questions regarding Northern Italy as a whole, starting from the dissolution of the Medieval communes, through to the rise of the signorie, from the end of the XIII century and the beginning of the XIV century up to the early XVI century.

Estudi estratègic del coblisme com a bé cultural

Aquest treball de final de carrera tracta sobre l’aplicació de les modernes tècniques de reflexió estratègica aplicades al sector de la cultura popular catalana, concretament la cultura popular catalana que engloba les manifestacions on el vincle comú és la formació musical pròpia: la cobla. Aprofitar aquest sector, a priori tant complex i tant allunyat en principi del món d’estratègia d’empresa com aquest, és l’excusa per veure el comportament d’aquestes eines i teories, i comprovar-ne la seva eficàcia i adaptabilitat als diferents casos de la realitat, a més de poder treballar un tema tant nou com la valoració d’intangibles. A la vegada, es produeix una síntesi de treballs orientats a l’organització d’entitats sense ànim de lucre existents en la bibliografia especialitzada, i que han tingut un inexistent ressò pràctic. Això pot ser un contrasentit en un país on l’associacionisme és tan arrelat i un dels pilars socials que ens vertebren i diferencien. En el cas de la música per a cobla, especialment destacable en un moment molt important i crític per estar en un temps de transició social molt fort. També, cal destacar vivament que té una sèrie de valors afegits de qualitat i complexitat que la situen en un interessant camí entre la cultura popular i les propostes de caràcter més comercial, en unes condicions actuals de gran desconeixement social. Les vinculacions entre estratègia, innovació, responsabilitat social i cultura del coneixement, donen a aquest treball un fort caràcter actual i adaptat a la realitat en les consideracions sobre el management en general.

Intercultural bridges in teenagers' theatrical events : performing self and constructing cultural identity through a creative drama process

The overriding aim of this drama educational case study is to deepen the understanding of meaning making in a creative intercultural youth theatre process and to examine it in the context of the 10th European Children's TheatreEncounter. The research task is to give a theoretical description of some key features of a creative drama process as the basis for theory about meaning makingin physical theatre. The first task is to illuminate the culture-historical connections of the multilayered practice of the EDERED-association. The second taskis to analyse and interpret theatrical meaning making. The ethnographical research site is regarded as a theatrical event. The analysis of the theatrical eventis divided into four segments: cultural contexts, contextual theatricality, theatrical playing and playing culture. These segments are connected with four research questions: What are the cultural contexts of a creative drama process? How can the organisation of the Encounter, genres, aesthetic codes and perception ofcodes be seen to influence the lived experiences of the participants? What are some of the key phases and characteristics in a creative practice? What kind of cultural learning can be interpreted from the performance texts? The interpretative question concerns identity and community (re)construction. How are the categories, `community´ and `child´ constructed in the Encounter culture? In this drama educational case study the research material (transcribed interviews, coded questionnaire answers, participant drawings, videotaped process text and performance texts) are examined in a multi-method analysis in the meta-theoretical framework of Dewey's naturalistic pragmatism. A three-dimensional research interest through a combination of lived experiences, social contexts and cultural-aesthetical practices compared with drama-educational practices required the methodological project of cultural studies. Furthermore, the critical interpretation of cultural texts is divided into three levels of analyses which are called description, structural analysis and theoretical interpretation. Dialogic validity (truthfulness, self-reflexivity and polyvocality) is combined with contextual validity (sensitivity to social context and awareness of historicity) and with deconstructive validity (awareness of the social discourses). My research suggests that itis possible, by means of physical theatre, to construct symbolic worlds where questions about intercultural identity and multilingual community are examined and where provisional answers are constructed in social interaction.