Quantifying the properties of interest is an important problem in many domains, e.g., assessing the condition of a patient, estimating the risk of an investment or relevance of the search result. However, the properties of interest are often latent and hard to assess directly, making it difficult to obtain classification or regression labels, which are needed to learn a predictive models from observable features. In such cases, it is typically much easier to obtain relative comparison of two instances, i.e. to assess which one is more intense (with respect to the property of interest). One framework able to learn from such kind of supervised information is ranking SVM, and it will make a basis of our approach. Applications in bio-medical datasets typically have specific additional challenges. First, and the major one, is the limited amount of data examples, due to an expensive measuring technology, and/or infrequency of conditions of interest. Such limited number of examples makes both identification of patterns/models and their validation less useful and reliable. Repeated samples from the same subject are collected on multiple occasions over time, which breaks IID sample assumption and introduces dependency structure that needs to be taken into account more appropriately. Also, feature vectors are highdimensional, and typically of much higher cardinality than the number of samples, making models less useful and their learning less efficient. Hypothesis of this dissertation is that use of the functional norm regularization can help alleviating mentioned challenges, by improving generalization abilities and/or learning efficiency of predictive models, in this case specifically of the approaches based on the ranking SVM framework. The temporal nature of data was addressed with loss that fosters temporal smoothness of functional mapping, thus accounting for assumption that temporally proximate samples are more correlated. Large number of feature variables was handled using the sparsity inducing L1 norm, such that most of the features have zero effect in learned functional mapping. Proposed sparse (temporal) ranking objective is convex but non-differentiable, therefore smooth dual form is derived, taking the form of quadratic function with box constraints, which allows efficient optimization. For the case where there are multiple similar tasks, joint learning approach based on matrix norm regularization, using trace norm L* and sparse row L21 norm was also proposed. Alternate minimization with proximal optimization algorithm was developed to solve the mentioned multi-task objective. Generalization potentials of the proposed high-dimensional and multi-task ranking formulations were assessed in series of evaluations on synthetically generated and real datasets. The high-dimensional approach was applied to disease severity score learning from gene expression data in human influenza cases, and compared against several alternative approaches. Application resulted in scoring function with improved predictive performance, as measured by fraction of correctly ordered testing pairs, and a set of selected features of high robustness, according to three similarity measures. The multi-task approach was applied to three human viral infection problems, and for learning the exam scores in Math and English. Proposed formulation with mixed matrix norm was overall more accurate than formulations with single norm regularization. / Computer and Information Science

Bacteria identification plays an essential role in preventing health complications and saving patients' lives. The most widely used method to identify bacteria, the bacterial cultural method, suffers from long processing times. Hence, an effective, rapid, and non-invasive method is needed as an alternative. Raman spectroscopy is a potential candidate for bacteria identifi cation due to its effective and rapid results and the fact that, similar to the uniqueness of a human fingerprint, the Raman spectrum is unique for every material. In my lab at the University of Ottawa, we focus on the use of Raman scattering for biosensing in order to achieve high identifi cation accuracy for different types of bacteria. Based on the unique Raman fingerprint for each bacteria type, different types of bacteria can be identifi ed successfully. However, using the Raman spectrum to identify bacteria poses a few challenges. First, the Raman signal is a weak signal, and so enhancement of the signal intensity is essential, e.g., by using surface-enhanced Raman scattering (SERS). Moreover, the Raman signal can be contaminated by different noise sources. Also, the signal consists of a large number of features, and is non-linear due to the correlation between the Raman features. Using machine learning (ML) along with SERS, we can overcome such challenges in the identifi cation process and achieve high accuracy for the system identifying bacteria. In this thesis, I present a method to improve the identifi cation of different bacteria types using a support vector machine (SVM) ML algorithm based on SERS. I also present dimension reduction techniques to reduce the complexity and processing time while maintaining high identifi cation accuracy in the classifi cation process. I consider four bacteria types: Escherichia coli (EC), Cutibacterium acnes (CA, it was formerly known as Propi-onibacterium acnes), methicillin-resistant Staphylococcus aureus (MRSA), and methicillin-sensitive Staphylococcus aureus (MSSA). Both the MRSA and MSSA are combined in a single class named MS in the classifi cation. We are focusing on using these types of bacteria as they are the most common types in the joint infection disease. Using binary classi fication, I present the simulation results for three binary models: EC vs CA, EC vs MS, and MS vs CA. Using the full data set, binary classi fication achieved a classi fication accuracy of more than 95% for the three models. When the samples data set was reduced, to decrease the complexity based on the samples' signal-to-noise ratio (SNR), a classi fication accuracy of more than 95% for the three models was achieved using less than 60% of the original data set. The recursive feature elimination (RFE) algorithm was then used to reduce the complexity in the feature dimension. Given that a small number of features were more heavily weighted than the rest of the features, the number of features used in the classifi cation could be signi ficantly reduced while maintaining high classi fication accuracy. I also present the classifi cation accuracy of using the multiclass one-versus-all (OVA) method, i.e., EC vs all, MS vs all, and CA vs all. Using the complete data set, the OVA method achieved classi cation accuracy of more than 90%. Similar to the binary classifi cation, the dimension reduction was applied to the input samples. Using the SNR reduction, the input samples were reduced by more than 60% while maintaining classifi cation accuracy higher than 80%. Furthermore, when the RFE algorithm was used to reduce the complexity on the features, and only the 5% top-weighted features of the full data set were used, a classi fication accuracy of more than 90% was achieved. Finally, by combining both reduction dimensions, the classi fication accuracy for the reduced data set was above 92% for a signifi cantly reduced data set. Both the dimension reduction and the improvement in the classi fication accuracy between different types of bacteria using the ML algorithm and SERS could have a signi ficant impact in ful lfiling the demand for accurate, fast, and non-destructive identi fication of bacteria samples in the medical fi eld, in turn potentially reducing health complications and saving patient lives.

Three-phase voltage source converters(VSCs) have become the converter of choice in many ac medium and high power applications due to their many advantages, including low harmonics, high power factor, and high efficiency. Modular VSCs have also been a popular choice as building blocks to achieve even higher power, primarily through converter paralleling. In addition to high power ratings, paralleling converters can also provide system redundancy through the so-called (N+1) configuration for improved availability, as well as allow easy implementation of converter power management. Interleaving can further improve the benefit of paralleling VSCs by reducing system harmonic currents, which potentially can increase system power density. There are many challenges to implement interleaving in paralleled VSCs system due to the complicated relationships in a three-phase power converter system. In addition, to maximize the benefit of interleaving, current knowledge of symmetric interleaving is not enough. More insightful understanding of this PWM technology is necessary before implement interleaving in a real paralleled VSCs system. In this dissertation, a systematic methodology to analyze and design a paralleled three-phase voltage source converters with interleaving is developed. All the analysis and proposed control methods are investigated with the goal of maximizing the benefit of interleaving based on system requirement. The dissertation is divided into five sections. Firstly, a complete analysis studying the impact of interleaving on harmonic currents in ac and dc side passive components for paralleled VSCs is presented. The analysis performed considers the effects of modulation index, pulse-width-modulation (PWM) schemes, interleaving angle and displacement angle. Based on the analysis the method to optimize interleaving angle is proposed. Secondly, the control methods for the common mode (CM) circulating current of paralleled three-phase VSCs with discontinuous space-vector modulation (DPWM) and interleaving are proposed. With the control methods, DPWM and interleaving, which is a desirable combination, but not considered possible, can be implemented together. In addition, the total flux of integrated inter-phase inductor to limit circulating current can be minimized. Thirdly, a 15 kW three phase ac-dc rectifier is built with SiC devices. With the technologies presented in this dissertation, the specific power density can be pushed more than 2kW/lb. Fourthly, the converter system with low switching frequency is studied. Special issues such as beat phenomenon and system unbalance due to non-triplen carrier ratio is explained and solved by control methods. Other than that, an improved asymmetric space vector modulation is proposed, which can significantly reduce output current total harmonic distortion (THD) for single and interleaved VSCs system. Finally, the method to protect a system with paralleled VSCs under the occurrence of internal faults is studied. After the internal fault is detected and isolated, the paralleled VSCs system can continue work. So system reliability can be increased. / Ph. D.

Power electronics building blocks (PEBBs) are standardized building blocks used to integrate power electronics systems. The PEBB approach can achieve low cost, high redundancy, high reliability, high flexibility and easy maintenance for large-scale power electronics systems. This thesis presents the implementation of a 100kW PEBB-based soft-switched bus regulator for an 800V DC distributed power system. The zero current transition (ZCT) soft-switching technique is used to improve the performance of the bus regulator by minimizing switching loss and improving overall efficiency. PEBB modules and a digital control building block are the subsystems of the DC bus regulator. This thesis addresses the design issues at subsystem and system levels. These include: operational principles and design of ZCT PEBB modules; design and implementation of the digital control block, based on DSP and EPLD; and modeling and control design of the DC bus regulator. There are several considerations when using the ZCT soft-switching technique in three-phase applications: the timing of the auxiliary switch gate signals must be arranged differently; there are low-frequency harmonics caused by the pulse width limits; and there is high thermal stress on the resonant capacitors. These issues are resolved by utilizing the sensed phase current information and the design freedom in the PWM modulator. A PWM modulation technique is proposed that can considerably reduce the switching events and further remove the associated loss while keeping THD low. Reduced switching events alleviate the thermal issue of the resonant capacitors. The same modulation technique can avoid the low-frequency harmonics caused by the pulse width limits and double the sampling frequency. The phase current information is used to deal with the control timing issue of the auxiliary switches and to control the three-phase soft-switching operation in order to achieve better efficiency. Additionally, the phase current information is used to implement dead time compensation to reduce THD. The soft-switched DC bus regulator has been tested up to a 100kW power level with 20kHz switching frequency. Experimental results demonstrate that high performance of the DC bus regulator is accomplished in terms of wide control bandwidth, low THD, unity power factor, high efficiency and high power density. / Master of Science

Water quality monitoring is crucial globally due to the vital role of freshwater in providing drinking water, irrigation, and ecosystem services. Highly polluted water poses risks to both ecosystems and human health. Current water quality monitoring methods deployed in the field are often expensive, labor-intensive, and invasive. To overcome these issues, this degree project investigated the use of remote sensing to assess critical water quality parameters in the Swedish river Indalsälven. The research questions focus on determining the accuracy of predicting chemical oxygen demand (COD), river color, turbidity, and total phosphorus (TP) using satellite data and machine learning algorithms. The findings revealed that COD can be predicted with a cross-validated coefficient of determination (R²CV) of 0.7, indicating a robust predictive capability. The study suggests that while approximate quantitative prediction of COD in oligotrophic rivers is feasible using Sentinel-2 imagery, predictions for the other parameters remain challenging in the context of Indalsälven. Improvements in prediction accuracy were achieved through optimized band combinations, reduced datasets encompassing satellite data collected within two days of field measurements, and suitable pre-processing methods. / Airborne Monitoring of Water Quality in Remote Regions

Le but de cette thèse est de développer des méthodes pratiques plus performantes pour la détection d'instances de classes d'objets de la vie quotidienne dans les images. Nous présentons une famille de détecteurs qui incorporent trois types d'indices visuelles performantes - histogrammes de gradients orientés (Histograms of Oriented Gradients, HOG), motifs locaux binaires (Local Binary Patterns, LBP) et motifs locaux ternaires (Local Ternary Patterns, LTP) - dans des méthodes de discrimination efficaces de type machine à vecteur de support latent (Latent SVM), sous deux régimes de réduction de dimension - moindres carrées partielles (Partial Least Squares, PLS) et sélection de variables par élagage de poids SVM (SVM Weight Truncation). Sur plusieurs jeux de données importantes, notamment ceux du PASCAL VOC2006 et VOC2007, INRIA Person et ETH Zurich, nous démontrons que nos méthodes améliorent l'état de l'art du domaine. Nos contributions principales sont : Nous étudions l'indice visuelle LTP pour la détection d'objets. Nous démontrons que sa performance est globalement mieux que celle des indices bien établies HOG et LBP parce qu'elle permet d'encoder à la fois la texture locale de l'objet et sa forme globale, tout en étant résistante aux variations d'éclairage. Grâce à ces atouts, LTP fonctionne aussi bien pour les classes qui sont caractérisées principalement par leurs structures que pour celles qui sont caractérisées par leurs textures. En plus, nous démontrons que les indices HOG, LBP et LTP sont bien complémentaires, de sorte qu'un jeux d'indices étendu qui intègre tous les trois améliore encore la performance. Les jeux d'indices visuelles performantes étant de dimension assez élevée, nous proposons deux méthodes de réduction de dimension afin d'améliorer leur vitesse et réduire leur utilisation de mémoire. La première, basée sur la projection moindres carrés partielles, diminue significativement le temps de formation des détecteurs linéaires, sans réduction de précision ni perte de vitesse d'exécution. La seconde, fondée sur la sélection de variables par l'élagage des poids du SVM, nous permet de réduire le nombre d'indices actives par un ordre de grandeur avec une réduction minime, voire même une petite augmentation, de la précision du détecteur. Malgré sa simplicité, cette méthode de sélection de variables surpasse toutes les autres approches que nous avons mis à l'essai.

Training a Support Vector Machine (SVM) requires the solution of a very large quadratic programming (QP) optimization problem. Sequential Minimal Optimization (SMO) is a decomposition-based algorithm which breaks this large QP problem into a series of smallest possible QP problems. However, it still costs O(n2) computation time. In our SVM implementation, we can do training with huge data sets in a distributed manner (by breaking the dataset into chunks, then using Message Passing Interface (MPI) to distribute each chunk to a different machine and processing SVM training within each chunk). In addition, we moved the kernel calculation part in SVM classification to a graphics processing unit (GPU) which has zero scheduling overhead to create concurrent threads. In this thesis, we will take advantage of this GPU architecture to improve the classification performance of SVM.

Thesis (MComm)--University of Stellenbosch, 2003. / Please refer to full text for abstract.

Land use transition in dryland ecosystems is one of the major driving forces to landscape change that directly impacts the welfare of humans. In this study, the support vector machine (SVM) classification algorithm and cross tabulation matrix analysis are used to identify systematic and random processes of change. The magnitude and prevailing signals of land use transitions are assessed taking into account net change and swap change. Moreover, spatiotemporal patterns and the relationship of precipitation and the Normalized Difference Vegetation Index (NDVI) are explored to evaluate landscape degradation. The assessment showed that 44% of net change and about 54% of total change occurred during the study period, with the latter being due to swap change. The conversion of over 39% of woodland to cropland accounts for the existence of the highest loss of valuable ecosystem of the region. The spatial relationship of NDVI and precipitation also showed R2 of below 0.5 over 55% of the landscape with no significant changes in the precipitation trend, thus representing an indicative symptom of land degradation. This in-depth analysis of random and systematic landscape change is crucial for designing policy intervention to halt woodland degradation in this fragile environment.

<p>This report examines the problem of real-time head pose estimation in low-resolution football footage. A method is presented for inferring the head pose using a combination of footage and knowledge of the locations of the football and players. An ensemble of randomized ferns is compared with a support vector machine for processing the footage, while a support vector machine performs pattern recognition on the location data. Combining the two sources of information outperforms either in isolation. The location of the football turns out to be an important piece of information.</p> / QC 20100707 / Capturing and Visualizing Large scale Human Action (ACTVIS)