Large-scale clustering through functional embedding

We present a new framework for large-scale data clustering. The main idea is to modify functional dimensionality reduction techniques to directly optimize over discrete labels using stochastic gradient descent. Compared to methods like spectral clustering our approach solves a single optimization problem, rather than an ad-hoc two-stage optimization approach, does not require a matrix inversion, can easily encode prior knowledge in the set of implementable functions, and does not have an ?out-of-sample? problem. Experimental results on both artificial and real-world datasets show the usefulness of our approach.

Recent advances in remote sensing image processing

Remote sensing image processing is nowadays a mature research area. The techniques developed in the field allow many real-life applications with great societal value. For instance, urban monitoring, fire detection or flood prediction can have a great impact on economical and environmental issues. To attain such objectives, the remote sensing community has turned into a multidisciplinary field of science that embraces physics, signal theory, computer science, electronics, and communications. From a machine learning and signal/image processing point of view, all the applications are tackled under specific formalisms, such as classification and clustering, regression and function approximation, image coding, restoration and enhancement, source unmixing, data fusion or feature selection and extraction. This paper serves as a survey of methods and applications, and reviews the last methodological advances in remote sensing image processing.

How early can we predict Alzheimer's disease using computational anatomy?

Computational anatomy with magnetic resonance imaging (MRI) is well established as a noninvasive biomarker of Alzheimer's disease (AD); however, there is less certainty about its dependency on the staging of AD. We use classical group analyses and automated machine learning classification of standard structural MRI scans to investigate AD diagnostic accuracy from the preclinical phase to clinical dementia. Longitudinal data from the Alzheimer's Disease Neuroimaging Initiative were stratified into 4 groups according to the clinical status-(1) AD patients; (2) mild cognitive impairment (MCI) converters; (3) MCI nonconverters; and (4) healthy controls-and submitted to a support vector machine. The obtained classifier was significantly above the chance level (62%) for detecting AD already 4 years before conversion from MCI. Voxel-based univariate tests confirmed the plausibility of our findings detecting a distributed network of hippocampal-temporoparietal atrophy in AD patients. We also identified a subgroup of control subjects with brain structure and cognitive changes highly similar to those observed in AD. Our results indicate that computational anatomy can detect AD substantially earlier than suggested by current models. The demonstrated differential spatial pattern of atrophy between correctly and incorrectly classified AD patients challenges the assumption of a uniform pathophysiological process underlying clinically identified AD.

Multidimensional Adaptations for Open Learning Management Systems

Our work is focused on alleviating the workload for designers of adaptive courses on the complexity task of authoring adaptive learning designs adjusted to specific user characteristics and the user context. We propose an adaptation platform that consists in a set of intelligent agents where each agent carries out an independent adaptation task. The agents apply machine learning techniques to support the user modelling for the adaptation process

Strategy selection: An introduction to the modeling challenge

Modeling the mechanisms that determine how humans and other agents choose among different behavioral and cognitive processes-be they strategies, routines, actions, or operators-represents a paramount theoretical stumbling block across disciplines, ranging from the cognitive and decision sciences to economics, biology, and machine learning. By using the cognitive and decision sciences as a case study, we provide an introduction to what is also known as the strategy selection problem. First, we explain why many researchers assume humans and other animals to come equipped with a repertoire of behavioral and cognitive processes. Second, we expose three descriptive, predictive, and prescriptive challenges that are common to all disciplines which aim to model the choice among these processes. Third, we give an overview of different approaches to strategy selection. These include cost‐benefit, ecological, learning, memory, unified, connectionist, sequential sampling, and maximization approaches. We conclude by pointing to opportunities for future research and by stressing that the selection problem is far from being resolved.

Interest rates mapping

The present study deals with the analysis and mapping of Swiss franc interest rates. Interest rates depend on time and maturity, defining term structure of the interest rate curves (IRC). In the present study IRC are considered in a two-dimensional feature space - time and maturity. Exploratory data analysis includes a variety of tools widely used in econophysics and geostatistics. Geostatistical models and machine learning algorithms (multilayer perceptron and Support Vector Machines) were applied to produce interest rate maps. IR maps can be used for the visualisation and pattern perception purposes, to develop and to explore economical hypotheses, to produce dynamic asset-liability simulations and for financial risk assessments. The feasibility of an application of interest rates mapping approach for the IRC forecasting is considered as well. (C) 2008 Elsevier B.V. All rights reserved.

What matters for predicting spatial distributions of trees: techniques, data, or species' characteristics?

Data characteristics and species traits are expected to influence the accuracy with which species' distributions can be modeled and predicted. We compare 10 modeling techniques in terms of predictive power and sensitivity to location error, change in map resolution, and sample size, and assess whether some species traits can explain variation in model performance. We focused on 30 native tree species in Switzerland and used presence-only data to model current distribution, which we evaluated against independent presence-absence data. While there are important differences between the predictive performance of modeling methods, the variance in model performance is greater among species than among techniques. Within the range of data perturbations in this study, some extrinsic parameters of data affect model performance more than others: location error and sample size reduced performance of many techniques, whereas grain had little effect on most techniques. No technique can rescue species that are difficult to predict. The predictive power of species-distribution models can partly be predicted from a series of species characteristics and traits based on growth rate, elevational distribution range, and maximum elevation. Slow-growing species or species with narrow and specialized niches tend to be better modeled. The Swiss presence-only tree data produce models that are reliable enough to be useful in planning and management applications.

Impingement jet cooling of end windings in a high-speed electric machine

The rotational speed of high-speed electric machines is over 15 000 rpm. These machines are compact in size when compared to the power rate. As a consequence, the heat fluxes are at a high level and the adequacy of cooling becomes an important design criterion. In the high-speed machines, the air gap between the stator and rotor is a narrow flow channel. The cooling air is produced with a fan and the flow is then directed to the air gap. The flow in the gap does not provide sufficient cooling for the stator end windings, and therefore additional cooling is required. This study investigates the heat transfer and flow fields around the coil end windings when cooling jets are used. As a result, an innovative and new assembly is introduced for the cooling jets, with the benefits of a reduced amount of hot spots, a lower pressure drop, and hence a lower power need for the cooling fan. The gained information can also be applied to improve the cooling of electric machines through geometry modifications. The objective of the research is to determine the locations of the hot spots and to find out induced pressure losses with different jet alternatives. Several possibilities to arrange the extra cooling are considered. In the suggested approach cooling is provided by using a row of air jets. The air jets have three main tasks: to cool the coils effectively by direct impingement jets, to increase and cool down the flow that enters the coil end space through the air gap, and to ensure the correct distribution of the flow by forming an air curtain with additional jets. One important aim of this study is the arrangement of cooling jets in such manner that hot spots can be avoided to wide extent. This enables higher power density in high-speed motors. This cooling system can also be applied to the ordinary electric machines when efficient cooling is needed. The numerical calculations have been performed using a commercial Computational Fluid Dynamics software. Two geometries have been generated: cylindrical for the studied machine and Cartesian for the experimental model. The main parameters include the positions, arrangements and number of jets, the jet diameters, and the jet velocities. The investigated cases have been tested with two widely used turbulence models and using a computational grid of over 500 000 cells. The experimental tests have been made by using a simplified model for the end winding space with cooling jets. In the experiments, an emphasis has been given to flow visualisation. The computational analysis shows good agreement with the experimental results. Modelling of the cooling jet arrangement enables also a better understanding of the complex system of heat transfer at end winding space.

2

The quality of environmental data analysis and propagation of errors are heavily affected by the representativity of the initial sampling design [CRE 93, DEU 97, KAN 04a, LEN 06, MUL07]. Geostatistical methods such as kriging are related to field samples, whose spatial distribution is crucial for the correct detection of the phenomena. Literature about the design of environmental monitoring networks (MN) is widespread and several interesting books have recently been published [GRU 06, LEN 06, MUL 07] in order to clarify the basic principles of spatial sampling design (monitoring networks optimization) based on Support Vector Machines was proposed. Nonetheless, modelers often receive real data coming from environmental monitoring networks that suffer from problems of non-homogenity (clustering). Clustering can be related to the preferential sampling or to the impossibility of reaching certain regions.

Sistema de presa de decisions distribuït

DDM is a framework that combines intelligent agents and artificial intelligence traditional algorithms such as classifiers. The central idea of this project is to create a multi-agent system that allows to compare different views into a single one.

CleanEx: a database of heterogeneous gene expression data based on a consistent gene nomenclature and linked to an improved annotation system

Résumé: L'automatisation du séquençage et de l'annotation des génomes, ainsi que l'application à large échelle de méthodes de mesure de l'expression génique, génèrent une quantité phénoménale de données pour des organismes modèles tels que l'homme ou la souris. Dans ce déluge de données, il devient très difficile d'obtenir des informations spécifiques à un organisme ou à un gène, et une telle recherche aboutit fréquemment à des réponses fragmentées, voir incomplètes. La création d'une base de données capable de gérer et d'intégrer aussi bien les données génomiques que les données transcriptomiques peut grandement améliorer la vitesse de recherche ainsi que la qualité des résultats obtenus, en permettant une comparaison directe de mesures d'expression des gènes provenant d'expériences réalisées grâce à des techniques différentes. L'objectif principal de ce projet, appelé CleanEx, est de fournir un accès direct aux données d'expression publiques par le biais de noms de gènes officiels, et de représenter des données d'expression produites selon des protocoles différents de manière à faciliter une analyse générale et une comparaison entre plusieurs jeux de données. Une mise à jour cohérente et régulière de la nomenclature des gènes est assurée en associant chaque expérience d'expression de gène à un identificateur permanent de la séquence-cible, donnant une description physique de la population d'ARN visée par l'expérience. Ces identificateurs sont ensuite associés à intervalles réguliers aux catalogues, en constante évolution, des gènes d'organismes modèles. Cette procédure automatique de traçage se fonde en partie sur des ressources externes d'information génomique, telles que UniGene et RefSeq. La partie centrale de CleanEx consiste en un index de gènes établi de manière hebdomadaire et qui contient les liens à toutes les données publiques d'expression déjà incorporées au système. En outre, la base de données des séquences-cible fournit un lien sur le gène correspondant ainsi qu'un contrôle de qualité de ce lien pour différents types de ressources expérimentales, telles que des clones ou des sondes Affymetrix. Le système de recherche en ligne de CleanEx offre un accès aux entrées individuelles ainsi qu'à des outils d'analyse croisée de jeux de donnnées. Ces outils se sont avérés très efficaces dans le cadre de la comparaison de l'expression de gènes, ainsi que, dans une certaine mesure, dans la détection d'une variation de cette expression liée au phénomène d'épissage alternatif. Les fichiers et les outils de CleanEx sont accessibles en ligne (http://www.cleanex.isb-sib.ch/). Abstract: The automatic genome sequencing and annotation, as well as the large-scale gene expression measurements methods, generate a massive amount of data for model organisms. Searching for genespecific or organism-specific information througout all the different databases has become a very difficult task, and often results in fragmented and unrelated answers. The generation of a database which will federate and integrate genomic and transcriptomic data together will greatly improve the search speed as well as the quality of the results by allowing a direct comparison of expression results obtained by different techniques. The main goal of this project, called the CleanEx database, is thus to provide access to public gene expression data via unique gene names and to represent heterogeneous expression data produced by different technologies in a way that facilitates joint analysis and crossdataset comparisons. A consistent and uptodate gene nomenclature is achieved by associating each single gene expression experiment with a permanent target identifier consisting of a physical description of the targeted RNA population or the hybridization reagent used. These targets are then mapped at regular intervals to the growing and evolving catalogues of genes from model organisms, such as human and mouse. The completely automatic mapping procedure relies partly on external genome information resources such as UniGene and RefSeq. The central part of CleanEx is a weekly built gene index containing crossreferences to all public expression data already incorporated into the system. In addition, the expression target database of CleanEx provides gene mapping and quality control information for various types of experimental resources, such as cDNA clones or Affymetrix probe sets. The Affymetrix mapping files are accessible as text files, for further use in external applications, and as individual entries, via the webbased interfaces . The CleanEx webbased query interfaces offer access to individual entries via text string searches or quantitative expression criteria, as well as crossdataset analysis tools, and crosschip gene comparison. These tools have proven to be very efficient in expression data comparison and even, to a certain extent, in detection of differentially expressed splice variants. The CleanEx flat files and tools are available online at: http://www.cleanex.isbsib. ch/.

Similarity measure based classification

Luokittelujärjestelmää suunniteltaessa tarkoituksena on rakentaa systeemi, joka pystyy ratkaisemaan mahdollisimman tarkasti tutkittavan ongelma-alueen. Hahmontunnistuksessa tunnistusjärjestelmän ydin on luokitin. Luokittelun sovellusaluekenttä on varsin laaja. Luokitinta tarvitaan mm. hahmontunnistusjärjestelmissä, joista kuvankäsittely toimii hyvänä esimerkkinä. Myös lääketieteen parissa tarkkaa luokittelua tarvitaan paljon. Esimerkiksi potilaan oireiden diagnosointiin tarvitaan luokitin, joka pystyy mittaustuloksista päättelemään mahdollisimman tarkasti, onko potilaalla kyseinen oire vai ei. Väitöskirjassa on tehty similaarisuusmittoihin perustuva luokitin ja sen toimintaa on tarkasteltu mm. lääketieteen paristatulevilla data-aineistoilla, joissa luokittelutehtävänä on tunnistaa potilaan oireen laatu. Väitöskirjassa esitetyn luokittimen etuna on sen yksinkertainen rakenne, josta johtuen se on helppo tehdä sekä ymmärtää. Toinen etu on luokittimentarkkuus. Luokitin saadaan luokittelemaan useita eri ongelmia hyvin tarkasti. Tämä on tärkeää varsinkin lääketieteen parissa, missä jo pieni tarkkuuden parannus luokittelutuloksessa on erittäin tärkeää. Väitöskirjassa ontutkittu useita eri mittoja, joilla voidaan mitata samankaltaisuutta. Mitoille löytyy myös useita parametreja, joille voidaan etsiä juuri kyseiseen luokitteluongelmaan sopivat arvot. Tämä parametrien optimointi ongelma-alueeseen sopivaksi voidaan suorittaa mm. evoluutionääri- algoritmeja käyttäen. Kyseisessä työssä tähän on käytetty geneettistä algoritmia ja differentiaali-evoluutioalgoritmia. Luokittimen etuna on sen joustavuus. Ongelma-alueelle on helppo vaihtaa similaarisuusmitta, jos kyseinen mitta ei ole sopiva tutkittavaan ongelma-alueeseen. Myös eri mittojen parametrien optimointi voi parantaa tuloksia huomattavasti. Kun käytetään eri esikäsittelymenetelmiä ennen luokittelua, tuloksia pystytään parantamaan.

Error models in hydrogeology applications

Notre consommation en eau souterraine, en particulier comme eau potable ou pour l'irrigation, a considérablement augmenté au cours des années. De nombreux problèmes font alors leur apparition, allant de la prospection de nouvelles ressources à la remédiation des aquifères pollués. Indépendamment du problème hydrogéologique considéré, le principal défi reste la caractérisation des propriétés du sous-sol. Une approche stochastique est alors nécessaire afin de représenter cette incertitude en considérant de multiples scénarios géologiques et en générant un grand nombre de réalisations géostatistiques. Nous rencontrons alors la principale limitation de ces approches qui est le coût de calcul dû à la simulation des processus d'écoulements complexes pour chacune de ces réalisations. Dans la première partie de la thèse, ce problème est investigué dans le contexte de propagation de l'incertitude, oú un ensemble de réalisations est identifié comme représentant les propriétés du sous-sol. Afin de propager cette incertitude à la quantité d'intérêt tout en limitant le coût de calcul, les méthodes actuelles font appel à des modèles d'écoulement approximés. Cela permet l'identification d'un sous-ensemble de réalisations représentant la variabilité de l'ensemble initial. Le modèle complexe d'écoulement est alors évalué uniquement pour ce sousensemble, et, sur la base de ces réponses complexes, l'inférence est faite. Notre objectif est d'améliorer la performance de cette approche en utilisant toute l'information à disposition. Pour cela, le sous-ensemble de réponses approximées et exactes est utilisé afin de construire un modèle d'erreur, qui sert ensuite à corriger le reste des réponses approximées et prédire la réponse du modèle complexe. Cette méthode permet de maximiser l'utilisation de l'information à disposition sans augmentation perceptible du temps de calcul. La propagation de l'incertitude est alors plus précise et plus robuste. La stratégie explorée dans le premier chapitre consiste à apprendre d'un sous-ensemble de réalisations la relation entre les modèles d'écoulement approximé et complexe. Dans la seconde partie de la thèse, cette méthodologie est formalisée mathématiquement en introduisant un modèle de régression entre les réponses fonctionnelles. Comme ce problème est mal posé, il est nécessaire d'en réduire la dimensionnalité. Dans cette optique, l'innovation du travail présenté provient de l'utilisation de l'analyse en composantes principales fonctionnelles (ACPF), qui non seulement effectue la réduction de dimensionnalités tout en maximisant l'information retenue, mais permet aussi de diagnostiquer la qualité du modèle d'erreur dans cet espace fonctionnel. La méthodologie proposée est appliquée à un problème de pollution par une phase liquide nonaqueuse et les résultats obtenus montrent que le modèle d'erreur permet une forte réduction du temps de calcul tout en estimant correctement l'incertitude. De plus, pour chaque réponse approximée, une prédiction de la réponse complexe est fournie par le modèle d'erreur. Le concept de modèle d'erreur fonctionnel est donc pertinent pour la propagation de l'incertitude, mais aussi pour les problèmes d'inférence bayésienne. Les méthodes de Monte Carlo par chaîne de Markov (MCMC) sont les algorithmes les plus communément utilisés afin de générer des réalisations géostatistiques en accord avec les observations. Cependant, ces méthodes souffrent d'un taux d'acceptation très bas pour les problèmes de grande dimensionnalité, résultant en un grand nombre de simulations d'écoulement gaspillées. Une approche en deux temps, le "MCMC en deux étapes", a été introduite afin d'éviter les simulations du modèle complexe inutiles par une évaluation préliminaire de la réalisation. Dans la troisième partie de la thèse, le modèle d'écoulement approximé couplé à un modèle d'erreur sert d'évaluation préliminaire pour le "MCMC en deux étapes". Nous démontrons une augmentation du taux d'acceptation par un facteur de 1.5 à 3 en comparaison avec une implémentation classique de MCMC. Une question reste sans réponse : comment choisir la taille de l'ensemble d'entrainement et comment identifier les réalisations permettant d'optimiser la construction du modèle d'erreur. Cela requiert une stratégie itérative afin que, à chaque nouvelle simulation d'écoulement, le modèle d'erreur soit amélioré en incorporant les nouvelles informations. Ceci est développé dans la quatrième partie de la thèse, oú cette méthodologie est appliquée à un problème d'intrusion saline dans un aquifère côtier. -- Our consumption of groundwater, in particular as drinking water and for irrigation, has considerably increased over the years and groundwater is becoming an increasingly scarce and endangered resource. Nofadays, we are facing many problems ranging from water prospection to sustainable management and remediation of polluted aquifers. Independently of the hydrogeological problem, the main challenge remains dealing with the incomplete knofledge of the underground properties. Stochastic approaches have been developed to represent this uncertainty by considering multiple geological scenarios and generating a large number of realizations. The main limitation of this approach is the computational cost associated with performing complex of simulations in each realization. In the first part of the thesis, we explore this issue in the context of uncertainty propagation, where an ensemble of geostatistical realizations is identified as representative of the subsurface uncertainty. To propagate this lack of knofledge to the quantity of interest (e.g., the concentration of pollutant in extracted water), it is necessary to evaluate the of response of each realization. Due to computational constraints, state-of-the-art methods make use of approximate of simulation, to identify a subset of realizations that represents the variability of the ensemble. The complex and computationally heavy of model is then run for this subset based on which inference is made. Our objective is to increase the performance of this approach by using all of the available information and not solely the subset of exact responses. Two error models are proposed to correct the approximate responses follofing a machine learning approach. For the subset identified by a classical approach (here the distance kernel method) both the approximate and the exact responses are knofn. This information is used to construct an error model and correct the ensemble of approximate responses to predict the "expected" responses of the exact model. The proposed methodology makes use of all the available information without perceptible additional computational costs and leads to an increase in accuracy and robustness of the uncertainty propagation. The strategy explored in the first chapter consists in learning from a subset of realizations the relationship between proxy and exact curves. In the second part of this thesis, the strategy is formalized in a rigorous mathematical framework by defining a regression model between functions. As this problem is ill-posed, it is necessary to reduce its dimensionality. The novelty of the work comes from the use of functional principal component analysis (FPCA), which not only performs the dimensionality reduction while maximizing the retained information, but also allofs a diagnostic of the quality of the error model in the functional space. The proposed methodology is applied to a pollution problem by a non-aqueous phase-liquid. The error model allofs a strong reduction of the computational cost while providing a good estimate of the uncertainty. The individual correction of the proxy response by the error model leads to an excellent prediction of the exact response, opening the door to many applications. The concept of functional error model is useful not only in the context of uncertainty propagation, but also, and maybe even more so, to perform Bayesian inference. Monte Carlo Markov Chain (MCMC) algorithms are the most common choice to ensure that the generated realizations are sampled in accordance with the observations. Hofever, this approach suffers from lof acceptance rate in high dimensional problems, resulting in a large number of wasted of simulations. This led to the introduction of two-stage MCMC, where the computational cost is decreased by avoiding unnecessary simulation of the exact of thanks to a preliminary evaluation of the proposal. In the third part of the thesis, a proxy is coupled to an error model to provide an approximate response for the two-stage MCMC set-up. We demonstrate an increase in acceptance rate by a factor three with respect to one-stage MCMC results. An open question remains: hof do we choose the size of the learning set and identify the realizations to optimize the construction of the error model. This requires devising an iterative strategy to construct the error model, such that, as new of simulations are performed, the error model is iteratively improved by incorporating the new information. This is discussed in the fourth part of the thesis, in which we apply this methodology to a problem of saline intrusion in a coastal aquifer.

Fiber flows in global paper and pulp industry. A panel data approach.

The purpose of this thesis is to study factors that explain the bilateral fiber trade flows. This is done by analyzing bilateral trade flows during 1990-2006. It will be studied also, whether there are differences between fiber types. This thesis uses a gravity model approach to study the trade flows. Gravity model is mostly used to study the aggregate data between trading countries. In this thesis the gravity model is applied to single fibers. This model is then applied to panel data set. Results from the regression show clearly that there are benefits in studying different fibers in separate. The effects differ considerably from each other. Furthermore, this thesis speaks for the existence of Linder’s effect in certain fiber types.

A Dependency Parsing Approach to Biomedical Text Mining

Biomedical research is currently facing a new type of challenge: an excess of information, both in terms of raw data from experiments and in the number of scientific publications describing their results. Mirroring the focus on data mining techniques to address the issues of structured data, there has recently been great interest in the development and application of text mining techniques to make more effective use of the knowledge contained in biomedical scientific publications, accessible only in the form of natural human language. This thesis describes research done in the broader scope of projects aiming to develop methods, tools and techniques for text mining tasks in general and for the biomedical domain in particular. The work described here involves more specifically the goal of extracting information from statements concerning relations of biomedical entities, such as protein-protein interactions. The approach taken is one using full parsing—syntactic analysis of the entire structure of sentences—and machine learning, aiming to develop reliable methods that can further be generalized to apply also to other domains. The five papers at the core of this thesis describe research on a number of distinct but related topics in text mining. In the first of these studies, we assessed the applicability of two popular general English parsers to biomedical text mining and, finding their performance limited, identified several specific challenges to accurate parsing of domain text. In a follow-up study focusing on parsing issues related to specialized domain terminology, we evaluated three lexical adaptation methods. We found that the accurate resolution of unknown words can considerably improve parsing performance and introduced a domain-adapted parser that reduced the error rate of theoriginal by 10% while also roughly halving parsing time. To establish the relative merits of parsers that differ in the applied formalisms and the representation given to their syntactic analyses, we have also developed evaluation methodology, considering different approaches to establishing comparable dependency-based evaluation results. We introduced a methodology for creating highly accurate conversions between different parse representations, demonstrating the feasibility of unification of idiverse syntactic schemes under a shared, application-oriented representation. In addition to allowing formalism-neutral evaluation, we argue that such unification can also increase the value of parsers for domain text mining. As a further step in this direction, we analysed the characteristics of publicly available biomedical corpora annotated for protein-protein interactions and created tools for converting them into a shared form, thus contributing also to the unification of text mining resources. The introduced unified corpora allowed us to perform a task-oriented comparative evaluation of biomedical text mining corpora. This evaluation established clear limits on the comparability of results for text mining methods evaluated on different resources, prompting further efforts toward standardization. To support this and other research, we have also designed and annotated BioInfer, the first domain corpus of its size combining annotation of syntax and biomedical entities with a detailed annotation of their relationships. The corpus represents a major design and development effort of the research group, with manual annotation that identifies over 6000 entities, 2500 relationships and 28,000 syntactic dependencies in 1100 sentences. In addition to combining these key annotations for a single set of sentences, BioInfer was also the first domain resource to introduce a representation of entity relations that is supported by ontologies and able to capture complex, structured relationships. Part I of this thesis presents a summary of this research in the broader context of a text mining system, and Part II contains reprints of the five included publications.