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.

Corridor Management Pilot Projects, 2004

Managing existing and newly constructed highway corridors has recently become a significant concern in many states, including Iowa. As urban land and land on the urban fringe develops, there is pressure to add features such as commercial driveways, at-grade public road intersections, and traffic signals to arterial highway routes that should primarily serve high-speed traffic. This diminishes the speed and traffic capacity of such roadways and can also cause significant safety issues. if mobility and safety are diminished, the value of the highway investment is diminished. Since a major highway corridor improvement may cost tens of millions of dollars or more, corridor management is as critical to preserving that investment as is more "hard side" management practices such as pavement or bridge management. Corridor management is a process that applies access management principles to highway corridors in an attempt to balance the competing needs of traffic service, safety, and support for land development. This project helped to identify routes that should be given high priority for corridor management. The pilot study in the form of two corridor management case studies provides an analytical process that can be replicated along the other Iowa commuting corridors using commonly available transportation and land use data resources. It also offers a general set of guidelines for the Iowa Department of Transportation to use in the development of its own comprehensive corridor management program.

Investigating the relationship between pollination strategies and the size-advantage model in zoophilous plants using the reproductive biology of Arum cylindraceum and other European Arum species as case studies

The size-advantage model (SAM) explains the temporal variation of energetic investment on reproductive structures (i.e. male and female gametes and reproductive organs) in long-lived hermaphroditic plants and animals. It proposes that an increase in the resources available to an organism induces a higher relative investment on the most energetically costly sexual structures. In plants, pollination interactions are known to play an important role in the evolution of floral features. Because the SAM directly concerns flower characters, pollinators are expected to have a strong influence on the application of the model. This hypothesis, however, has never been tested. Here, we investigate whether the identity and diversity of pollinators can be used as a proxy to predict the application of the SAM in exclusive zoophilous plants. We present a new approach to unravel the dynamics of the model and test it on several widespread Arum (Araceae) species. By identifying the species composition, abundance and spatial variation of arthropods trapped in inflorescences, we show that some species (i.e. A. cylindraceum and A. italicum) display a generalist reproductive strategy, relying on the exploitation of a low number of dipterans, in contrast to the pattern seen in the specialist A. maculatum (pollinated specifically by two fly species only). Based on the model presented here, the application of the SAM is predicted for the first two and not expected in the latter species, those predictions being further confirmed by allometric measures. We here demonstrate that while an increase in the female zone occurs in larger inflorescences of generalist species, this does not happen in species demonstrating specific pollinators. This is the first time that this theory is both proposed and empirically tested in zoophilous plants. Its overall biological importance is discussed through its application in other non-Arum systems.

Assigning and combining probabilities in single-case studies

There is currently a considerable diversity of quantitative measures available for summarizing the results in single-case studies. Given that the interpretation of some of them is difficult due to the lack of established benchmarks, the current paper proposes an approach for obtaining further numerical evidence on the importance of the results, complementing the substantive criteria, visual analysis, and primary summary measures. This additional evidence consists of obtaining the statistical significance of the outcome when referred to the corresponding sampling distribution. This sampling distribution is formed by the values of the outcomes (expressed as data nonoverlap, R-squared, etc.) in case the intervention is ineffective. The approach proposed here is intended to offer the outcome"s probability of being as extreme when there is no treatment effect without the need for some assumptions that cannot be checked with guarantees. Following this approach, researchers would compare their outcomes to reference values rather than constructing the sampling distributions themselves. The integration of single-case studies is problematic, when different metrics are used across primary studies and not all raw data are available. Via the approach for assigning p values it is possible to combine the results of similar studies regardless of the primary effect size indicator. The alternatives for combining probabilities are discussed in the context of single-case studies pointing out two potentially useful methods one based on a weighted average and the other on the binomial test.

Strategic supply lead time management. Case Nokia Networks, Hardware Services

Teollisuuden palveluiden on huomattu olevan potentiaalinen lisätulojen lähde. Teollisuuden palveluiden dynaamisessa maailmassa räätälöinti ja kyky toimia nopeasti ovat kriittisiä asiakastyytyväisyyden ja kilpailuedun luomisprosessin osia. Toimitusketjussa käytetyn ajan lyhentämisellä voidaan saavuttaa sekä paremmat vasteajat, että alhaisemmat kokonaiskustannukset. Tutkielman tavoitteena on kuvata teollisuuden palveluiden dynaamista ympäristöä: asiakastarvetta, sekä mahdollisuuksia kaventaa pyydetyn ja saavutetun toimitusajan välistä eroa. Tämä toteutetaan pääosin strategisen toimitusajan hallinnan keinoin. Langattomien tietoliikenneverkkojen operaattorit haluavat vähentää ydinosaamiseensa kuulumatomiin toimintoihin, kuten ylläpitoon sitoutuneita pääomia. Tutkielman case osiossa varaosapalvelujen toimitusketjun kysyntä-, materiaali- ja informaatiovirtoja analysoidaan niin kvalitatiivisten haastatteluiden, sisäisten dokumenttien, kuin kvantitatiivisten tilastollisten menetelmienkin avulla. Löydöksiä peilataan vallitsevaa toimitusketjun ja ajanhallinnan paradigmaa vasten. Tulokset osoittavat, että vahvan palvelukulttuurin omaksuminen ja kokonaisvaltainen toimitusketjun tehokkuuden mittaaminen ovat ajanhallinnan lähtökohtia teollisuuden palveluissa.

Describing and Analyzing Information Technology through Total Quality Management and Process Management: Case Stora Enso Newsprint

Laatu on osaltaan vahvistamassa asemaansa liike-elämässä yritysten kilpaillessa kansainvälisillä markkinoilla niin hinnalla kuin laadulla. Tämä suuntaus on synnyttänyt useita laatuohjelmia, joita käytetään ahkerasti yritysten kokonais- valtaisen laatujohtamisen (TQM) toteuttamisessa. Laatujohtaminen kattaa yrityksen kaikki toiminnot ja luo vaatimuksia myös yrityksen tukitoimintojen kehittämiselle ja parantamiselle. Näihin lukeutuu myös tämän tutkimuksen kohde tietohallinto (IT). Tutkielman tavoitteena oli kuvata IT prosessin nykytila. Tutkielmassa laadittu prosessikuvaus pohjautuu prosessijohtamisen teoriaan ja kohdeyrityksen käyttämään laatupalkinto kriteeristöön. Tutkimusmenetelmänä prosessin nykytilan selvittämiseksi käytettiin teemahaastattelutta. Prosessin nykytilan ja sille asetettujen vaatimusten selvittämiseksi haastateltiin IT prosessin asiakkaita. Prosessianalyysi, tärkeimpien ala-prosessien tunnistaminen ja parannusalueiden löytäminen ovat tämän tutkielman keskeisemmät tulokset. Tutkielma painottui IT prosessin heikkouksien ja parannuskohteiden etsimiseen jatkuvan kehittämisen pohjaksi, ei niinkään prosessin radikaaliin uudistamiseen. Tutkielmassa esitellään TQM:n periaatteet, laatutyökaluja sekä prosessijohtamisen terminologia, periaatteet ja sen systemaattinen toteutus. Työ antaa myös kuvan siitä, miten TQM ja prosessijohtaminen niveltyvät yrityksen laatutyössä.

Selecting the most relevant variables for anaerobic digestion imbalances : Two case studies

In this study, a wrapper approach was applied to objectively select the most important variables related to two different anaerobic digestion imbalances, acidogenic states and foaming. This feature selection method, implemented in artificial neural networks (ANN), was performed using input and output data from a fully instrumented pilot plant (1 m 3 upflow fixed bed digester). Results for acidogenic states showed that pH, volatile fatty acids, and inflow rate were the most relevant variables. Results for foaming showed that inflow rate and total organic carbon were among the relevant variables, both of which were related to the feed loading of the digester. Because there is not a complete agreement on the causes of foaming, these results highlight the role of digester feeding patterns in the development of foaming

Least squares regression with errors in both variables: case studies

Analytical curves are normally obtained from discrete data by least squares regression. The least squares regression of data involving significant error in both x and y values should not be implemented by ordinary least squares (OLS). In this work, the use of orthogonal distance regression (ODR) is discussed as an alternative approach in order to take into account the error in the x variable. Four examples are presented to illustrate deviation between the results from both regression methods. The examples studied show that, in some situations, ODR coefficients must substitute for those of OLS, and, in other situations, the difference is not significant.

Understanding Neolithic Southern Levant: case studies of archaeological semiosis in action

I undersökningen tillämpas Charles Sanders Peirces semiotik för en kritisk granskning av arkeologiska tolkningsprocesser. Enligt Peirce bygger all betydelsegivning på tecken, som kan vara teckningar, föremål, ord, byggkonstruktioner eller egentligen vad som helst. Ett tecken är tredimensionellt: ”objekt”, ”tecken” och ”tolkning”. I sina tidiga skrifter definierar han tre grundtyper för tecknet, Index, Icon och Symbol. De grundläggande definitionerna i Peirces semiotik blir till ett slags lins. När den placeras på skrifter av en arkeolog som uttolkar tecken, framträder deras inre uppbyggnad, motiveringar och logiska konsekvens klart. Att beakta är, att denna bok är lika lite avsedd att utgöra en systematisk klarläggning av den arkeologiska semiotiken, som en omfattande beskrivning av symboliken i det neolitiska Mellanöstern. Analysen är deskriptiv och inte avsedd att utvärdera tolkningarnas riktighet, utan enbart att klarlägga hur arkeologen kommit fram till dessa. Som objekt har valts neolitiska södra Levanten, där viktiga fynd gällande denna i mänsklighetens kulturhistoria så betydelsefulla skede har gjorts. Förhistorien är intressant med tanke på arkeologisk semiosis, eftersom uttolkaren av en symbol inte kan stöda sig på textfynd, utan måste på annat sätt upptäcka vad ett föremål eller en byggnad betytt för sin upphovsman. Att upptäcka en trovärdig betydelse är ofta en mycket svår och understundom rentav omöjlig uppgift. Efter att förhållandet mellan semiotik och arkeologi dryftats analyseras i boken John Garstangs och Kathleen M. Kenyons grundläggande tolkningar i Jeriko, Denise-Schmandt Besserats jämförande analyser för uttolkningen av \'Ain Ghazalis kranium, Michele A. Millers kontextuella analys i Jarmuk samt David Lewis-Williams’ starkt strukturalistiska analys av betydelsen av fynden i \'Ain Ghazal. Peirces semiotik har använts som stöd för arkeologin i mycket mindre utsträckning än F. de Saussures lingvistiska ja strukturalistiska semiotik. I Mellanöstern har man hittills inte alls gjort det. Ingen av de forskare som behandlas i boken hänvisar själv till semiotik eller tecken. Logiken i uttolkningen av dessa undersökningar är mycket invecklad, och de av Peirce gestaltade processerna för betydelsegivning visar sig härvidlag utgöra en ytterst klargörande kritisk apparat.

Process approach in quality management - Case: ECO process development in EMS industry

The aim of the thesis was to study quality management with process approach and to find out how to utilize process management to improve quality. The operating environment of organizations has changed. Organizations are focusing on their core competences and networking with suppliers and customers to ensure more effective and efficient value creation for the end customer. Quality management is moving from inspection of the output to prevention of problems from occurring in the first place and management thinking is changing from functional approach to process approach. In the theoretical part of the thesis, it is studied how to define quality, how to achieve good quality, how to improve quality, and how to make sure the improvement goes on as never ending cycle. A selection of quality tools is introduced. Process approach to quality management is described and compared to functional approach, which is the traditional way to manage operations and quality. The customer focus is also studied, and it is presented, that to ensure long term customer commitment, organization needs to react to changing customer requirements and wishes by constantly improving the processes. In the experimental part the theories are tested in a process improvement business case. It is shown how to execute a process improvement project starting from defining the customer requirements, continuing to defining the process ownership, roles and responsibilities, boundaries, interfaces and the actual process activities. The control points and measures are determined for the process, as well as the feedback and corrective action process, to ensure continual improvement can be achieved and to enable verification that customer requirements are fulfilled.

“THE PATIENT SURVIVED DESPITE THE TREATMENT”. A study of Finnish health policy and New Public Management. Case: The Government Resolution on the Health 2015 public health programme.

Internationally, Finland has been among the most respected countries during several decades in terms of public health. WHO has had the most significant influence on Finnish health policy and the relationship has traditionally been warm. However, the situation has slightly changed in the last 10-20 years. The objectives of Finnish national health policy have been to secure the best possible health for the population and to minimize disparities in health between different population groups. Nevertheless, although the state of public health and welfare has steadily improved, the socioeconomic disparities in health have increased. This qualitative case study will demonstrate why health is political and why health matters. It will also present some recommendations for research topics and administrative reforms. It will be argued that lack of political interest in health policy leads to absence of health policy visions and political commitment, which can be disastrous for public health. This study will investigate how Finnish health policy is defined and organised, and it will also shed light on Finnish health policy formation processes and actors. Health policy is understood as a broader societal construct covering the domains of different ministries, not just Ministry of Social Affairs and Health (MSAH). The influences of economic recession of the 1990s, state subsidy reform in 1993, globalisation and the European Union will be addressed, as well. There is not much earlier Finnish research done on health policy from political science viewpoint. Therefore, this study is interdisciplinary and combines political science with administrative science, contemporary history and health policy research with a hint of epidemiology. As a method, literature review, semi-structured interviews and policy analysi will be utilised. Institutionalism, policy transfer, and corporatism are understood as the theoretical framework. According to the study, there are two health policies in Finland: the official health policy and health policy generated by industry, media and various interest organisations. The complex relationships between the Government and municipalities, and on the other hand, the MSAH and National Institute for Health and Welfare (THL) seemed significant in terms of Finnish health policy coordination. The study also showed that the Investigated case, Health 2015, does not fulfil all necessary criteria for a successful public health programme. There were also several features both in Health 2015 and Finnish health policy, which can be interpreted in NPM framework and seen having NPM influences.

Data management as a part of performance management: Case study about production reporting development project

Because of the increased availability of different kind of business intelligence technologies and tools it can be easy to fall in illusion that new technologies will automatically solve the problems of data management and reporting of the company. The management is not only about management of technology but also the management of processes and people. This thesis is focusing more into traditional data management and performance management of production processes which both can be seen as a requirement for long lasting development. Also some of the operative BI solutions are considered in the ideal state of reporting system. The objectives of this study are to examine what requirements effective performance management of production processes have for data management and reporting of the company and to see how they are effecting on the efficiency of it. The research is executed as a theoretical literary research about the subjects and as a qualitative case study about reporting development project of Finnsugar Ltd. The case study is examined through theoretical frameworks and by the active participant observation. To get a better picture about the ideal state of reporting system simple investment calculations are performed. According to the results of the research, requirements for effective performance management of production processes are automation in the collection of data, integration of operative databases, usage of efficient data management technologies like ETL (Extract, Transform, Load) processes, data warehouse (DW) and Online Analytical Processing (OLAP) and efficient management of processes, data and roles.

Facing the Hydra alone: three case studies

Presentation at Open Repositories 2014, Helsinki, Finland, June 9-13, 2014