Frequent itemset mining is an important building block in many data mining applications like market basket analysis, recommendation, web-mining, fraud detection, and gene expression analysis. In many of them, the datasets being mined can easily grow up to hundreds of gigabytes or even terabytes of data. Hence, efficient algorithms are required to process such large amounts of data. In recent years, there have been many frequent-itemset mining algorithms proposed, which however (1) often have high memory requirements and (2) do not exploit the large degrees of parallelism provided by modern multiprocessor systems. The high memory requirements arise mainly from inefficient data structures that have only been shown to be sufficient for small datasets. For large datasets, however, the use of these data structures force the algorithms to go out-of-core, i.e., they have to access secondary memory, which leads to serious performance degradations. Exploiting available parallelism is further required to mine large datasets because the serial performance of processors almost stopped increasing. Algorithms should therefore exploit the large number of available threads and also the other kinds of parallelism (e.g., vector instruction sets) besides thread-level parallelism. In this work, we tackle the high memory requirements of frequent itemset mining twofold: we (1) compress the datasets being mined because they must be kept in main memory during several mining invocations and (2) improve existing mining algorithms with memory-efficient data structures. For compressing the datasets, we employ efficient encodings that show a good compression performance on a wide variety of realistic datasets, i.e., the size of the datasets is reduced by up to 6.4x. The encodings can further be applied directly while loading the dataset from disk or network. Since encoding and decoding is repeatedly required for loading and mining the datasets, we reduce its costs by providing parallel encodings that achieve high throughputs for both tasks. For a memory-efficient representation of the mining algorithms’ intermediate data, we propose compact data structures and even employ explicit compression. Both methods together reduce the intermediate data’s size by up to 25x. The smaller memory requirements avoid or delay expensive out-of-core computation when large datasets are mined. For coping with the high parallelism provided by current multiprocessor systems, we identify the performance hot spots and scalability issues of existing frequent-itemset mining algorithms. The hot spots, which form basic building blocks of these algorithms, cover (1) counting the frequency of fixed-length strings, (2) building prefix trees, (3) compressing integer values, and (4) intersecting lists of sorted integer values or bitmaps. For all of them, we discuss how to exploit available parallelism and provide scalable solutions. Furthermore, almost all components of the mining algorithms must be parallelized to keep the sequential fraction of the algorithms as small as possible. We integrate the parallelized building blocks and components into three well-known mining algorithms and further analyze the impact of certain existing optimizations. Our algorithms are already single-threaded often up an order of magnitude faster than existing highly optimized algorithms and further scale almost linear on a large 32-core multiprocessor system. Although our optimizations are intended for frequent-itemset mining algorithms, they can be applied with only minor changes to algorithms that are used for mining of other types of itemsets.

Le rôle que pourrait jouer l’interaction des systèmes cholinergiques (ACh) et dopaminergiques (DA) semble crucial dans la physiopathologie de certaines maladies neurodégénératives, en particulier dans la démence à corps de Lewy (DCL). Ce travail de thèse se propose de valider un protocole d’imagerie moléculaire en tomographie d’émission monophotonique, consistant en un marquage de l’ACh et de la DA chez un même individu, afin de pouvoir étudier in vivo les interactions ACh/DA.Après avoir mis au point chez des sujets âgés et des patients atteints de maladie d’Alzheimer une méthode d’imagerie cérébrale quantitative des neurones ACh utilisant un radioligand sélectif du transporteur vésiculaire de l’ACh, le [123I]-IBVM, et basée sur une modélisation pharmacocinétique, nous avons montré le potentiel de cette méthode à mettre en évidence une atteinte différentielle des circuits ACh dans la Paralysie Supranucléaire Progressive et l’Atrophie Multisystématisée. Dans la dernière partie de ce travail de thèse, nous avons pour la première fois réalisé un double marquage des systèmes ACh et DA dans la DCL, en utilisant, en plus du [123I]-IBVM, un radioligand sélectif du transporteur de la dopamine et validé en routine clinique, le [123I]-FP-CIT. En parallèle, une étude comportementale évaluant la présence d’hallucinations, de fluctuations cognitives, d’altérations des rythmes circadiens ainsi qu’un bilan des performances neuropsychologiques, ont été menés. Cette étude est actuellement en cours de réalisation. Les tous premiers résultats montrent l’existence de liens cohérents entre les données d’imagerie moléculaire et les données cliniques. / The question of how acetylcholine (ACh) and dopamine (DA) could be involved together in the pathophysiology of some neurodegenerative disorders is essential, particularly in dementia with Lewy bodies (DLB). The present study aims at assessing an in vivo molecular imaging method of both ACh and DA brain systems using single photon emission computed tomography. In the first part of the present study, a method based on pharmacokinetic analysis making it possible to quantify ACh neurons in vivo, using [123I]-IBVM, a specific radioligand of vesicular acetylcholine transporter, was developed and validated in healthy subjects and Alzheimer’s disease patients. Then, we showed the ability of our method to demonstrate a differential alteration of ACh pathways in Progressive Supranuclear Palsy and Multiple System Atrophy patients. In the last part of this study, we imaged for the first time both ACh and DA systems in DLB patients, using not only [123I]-IBVM, but also [123I]-FP-CIT, a specific radioligand of dopamine transporter. Concomitantly, a behavioral exploration of hallucinations, fluctuating cognition and disturbances of circadian rhythms was achieved in these patients, as well as a neuropsychological examination. This study is currently in progress. The first results show consistent links between imaging and clinical data.

Frequent itemset mining is an important building block in many data mining applications like market basket analysis, recommendation, web-mining, fraud detection, and gene expression analysis. In many of them, the datasets being mined can easily grow up to hundreds of gigabytes or even terabytes of data. Hence, efficient algorithms are required to process such large amounts of data. In recent years, there have been many frequent-itemset mining algorithms proposed, which however (1) often have high memory requirements and (2) do not exploit the large degrees of parallelism provided by modern multiprocessor systems. The high memory requirements arise mainly from inefficient data structures that have only been shown to be sufficient for small datasets. For large datasets, however, the use of these data structures force the algorithms to go out-of-core, i.e., they have to access secondary memory, which leads to serious performance degradations. Exploiting available parallelism is further required to mine large datasets because the serial performance of processors almost stopped increasing. Algorithms should therefore exploit the large number of available threads and also the other kinds of parallelism (e.g., vector instruction sets) besides thread-level parallelism. In this work, we tackle the high memory requirements of frequent itemset mining twofold: we (1) compress the datasets being mined because they must be kept in main memory during several mining invocations and (2) improve existing mining algorithms with memory-efficient data structures. For compressing the datasets, we employ efficient encodings that show a good compression performance on a wide variety of realistic datasets, i.e., the size of the datasets is reduced by up to 6.4x. The encodings can further be applied directly while loading the dataset from disk or network. Since encoding and decoding is repeatedly required for loading and mining the datasets, we reduce its costs by providing parallel encodings that achieve high throughputs for both tasks. For a memory-efficient representation of the mining algorithms’ intermediate data, we propose compact data structures and even employ explicit compression. Both methods together reduce the intermediate data’s size by up to 25x. The smaller memory requirements avoid or delay expensive out-of-core computation when large datasets are mined. For coping with the high parallelism provided by current multiprocessor systems, we identify the performance hot spots and scalability issues of existing frequent-itemset mining algorithms. The hot spots, which form basic building blocks of these algorithms, cover (1) counting the frequency of fixed-length strings, (2) building prefix trees, (3) compressing integer values, and (4) intersecting lists of sorted integer values or bitmaps. For all of them, we discuss how to exploit available parallelism and provide scalable solutions. Furthermore, almost all components of the mining algorithms must be parallelized to keep the sequential fraction of the algorithms as small as possible. We integrate the parallelized building blocks and components into three well-known mining algorithms and further analyze the impact of certain existing optimizations. Our algorithms are already single-threaded often up an order of magnitude faster than existing highly optimized algorithms and further scale almost linear on a large 32-core multiprocessor system. Although our optimizations are intended for frequent-itemset mining algorithms, they can be applied with only minor changes to algorithms that are used for mining of other types of itemsets.

This Master's thesis deals with the principles of the data mining process, especially with the mining  of association rules. The theoretical apparatus of general description and principles of the data warehouse creation is set. On the basis of this theoretical knowledge, the application for the association rules mining is implemented. The application requires the data in the transactional form or the multidimensional data organized in the Star schema. The implemented algorithms for finding  of the frequent patterns are Apriori and FP-tree. The system allows the variant setting of parameters for mining process. Also, the validation tests and efficiency proofs were accomplished. From the point of view of the association rules searching support, the resultant application is more applicable and robust than the existing compared systems SAS Miner and Oracle Data Miner.

Flooding is of great concern world-wide, causing damage to infrastructure, property and loss of life. Low-income countries, in particular, can be negatively affected by flood events due to their inherent vulnerabilities. Moreover, data to perform studies for flood risk management in low-income regions are often scarce or lacking sufficient quality. This thesis proposes new methodologies and explores the use of unconventional sources of information in flood hazard assessment in areas where the quantity or sufficient quality of traditional hydrometrical data are lacking.  One method was developed to account for errors in spatially averaged rainfall, from a sparse rain-gauge network, used as input to a rainfall-runoff model. A spatially-averaged and event-dependent rainfall depth multiplier led to improvements of the hydrographs at calibration. And by using a distribution of the multiplier, identified from previous events in the catchment, improvement in predictions could also be obtained. A second method explored the possibility of reproducing an unmeasured extreme flood event using a combination of models, post-event data, precipitation and an uncertainty-analysis framework. This combination allowed the identification of likelihood-associated parameter sets from which the flood hazard map for the extreme event could be obtained. A third and fourth study made at the regional scale explored the value of catchment similarities, and the effects of climate on the hydrological response of catchments. Flood frequency curves were estimated for 36 basins, assumed ungauged, using regional information of short flow records, and local information about the frequency of the storm. In the second regional study, hydro-climatic information provided great value to constrain predictions of series of daily flow from a hydrological model. Previously described methods, used in combination with unconventional information within an uncertainty analysis, proven to be useful for flood hazard assessment at basins with data limitations. The explored data included: post-event measurements of an extreme flood event, hydro-climate regional information and local precipitation data. The methods presented in this thesis are expected to support development of hydrological studies underpinning flood-risk reduction in data-poor areas. / Extremt höga vattenflöden ställer till stora problem i hela världen. De skadar infrastruktur och egendom och orsakar död. Framför allt kan låg- och medelinkomstländer vara väldigt sårbara för extrema flöden. I dessa länder saknas dessutom ofta data som behövs för att kunna bedöma översvämningsrisker, eller så finns bara data av dålig kvalitet. Denna avhandling föreslår nya metoder som använder okonventionella informationskällor vid bedömning av översvämningsrisker i områden där traditionella hydrologiska data saknas eller har otillräcklig kvalitet. En metod utvecklades för att ta hänsyn till fel i rumslig medelnederbörd beräknad från ett glest nät av nederbördsmätare att användas som indata i en nederbörds-avrinningsmodell. Användning av en multiplikator för medelvärdesbildad nederbörd, i tid och rum, för enskilda högflödestillfällen ledde till förbättrad modellkalibrering. Genom att använda multiplikatorfördelningar, identifierade från tidigare högflödestillfällen i avrinningsområdet, kunde också prognoser förbättras. En andra metod använde sig av möjligheten att reproducera ett extremt högflöde inom ramen för en osäkerhetsanalys med hjälp av en kombination av modeller, nederbördsdata och data som uppmätts i efterhand. Denna kombination gjorde det möjligt att identifiera parametervärdesuppsättningar med hophörande sannolikheter ur vilka det gick att erhålla en översvämningskarta för det höga flödet. En tredje och fjärde studie i regional skala utforskade värdet av likheter mellan avrinningsområden och hur områdenas hydrologiska gensvar beror av klimatet. Kurvan för kumulativa högflödesfrekvenser (flood frequency curve, FFC) kunde skattas med hjälp av lokal nederbördsinformation och regional information om korta tidsserier av vattenföring från 36 avrinningsområden som antogs sakna vattenföringsdata. I den andra regionala studien visade sig hydroklimatisk information av värde för att avgränsa godtagbara prognoser för daglig vattenföring från en hydrologisk modell. Tidigare beskrivna metoder, använda tillsammans med okonventionell information inom ramen för en osäkerhetsanalys, visade sig vara användbara för att bedöma översvämningsrisker i avrinningsområden med databegränsningar. Bland utforskade data fanns: mätningar i efterhand av ett extremt högflöde, hydroklimatisk regional information och lokala nederbördsmätningar. Metoderna i denna avhandling förväntas kunna stödja utvecklingen av hydrologiska studier av höga flöden och översvämningar i områden med bristande datatillgång. / Las inundaciones ocasionan daños a la infraestructura, propiedad y pérdida de vidas a nivel mundial. Los países en desarrollo son los más vulnerables a inundaciones, la calidad y cantidad de datos hidro-climatológicos disponibles en los mismos dificulta el desarrollo de estudios para la evaluación de riesgo a esta amenaza. Esta tesis propone métodos en la que se hace uso de fuentes de información no-convencionales para la evaluación de riesgo por inundación en regiones con datos escasos o limitados. Un método considera el error asociado a la precipitación promedio sobre cuencas en modelos lluvia-escorrentía como un factor multiplicador del histograma del evento. El uso de la precipitación promedio junto con una distribución probabilística del factor multiplicador como datos de entrada a un modelo de lluvia-escorrentía mejoraron los hidrogramas durante los periodos de calibración y predicción. Un segundo método exploró la posibilidad de reproducir un evento extremo de inundación usando una combinación de modelos hidrológicos e hidráulico, un análisis de incertidumbre, datos hidrométricos recopilados después del evento y datos de precipitación registrados durante-el-evento. Dicha combinación permitió la identificación de los parámetros de los modelos y la elaboración un mapa de amenaza por inundaciones para dicho evento. Adicionalmente, se estimaron curvas de frecuencia de inundaciones para 36 cuencas, asumidas no aforadas, mediante un método de regionalización que usa registros de caudal de corta duración disponibles en la región. Dichas curvas fueron extendidas haciendo uso de información local sobre la frecuencia de las tormentas. Se encontró que la información hidro-climatológica tiene un gran valor para reducir el rango de incertidumbre de las simulaciones de caudal diaria de un modelo hidrológico. Los métodos anteriores se usaron en combinación con información no-convencional dentro de un análisis de incertidumbre y han probado su utilidad para la evaluación de riesgo por inundaciones en cuencas con registros escasos o limitados. Los datos utilizados en esta tesis incluyen datos hidrométricos recopilados pasado el evento, registros hidro-climatológicos regionales y precipitación local. Se espera que los métodos presentados aquí contribuyan al desarrollo de estudios hidrológicos importantes para la reducción del riesgo por inundaciones en regiones con déficit de registros hidro-climatológicos.

Cette thèse contribue à la compréhension fondamentale du phénomène de couplage fort de la lumière avec des molécules organiques en mettant en œuvre de nouveaux systèmes et de nouvelles techniques, afin d'étudier les modifications de propriétés de molécules couplées à des résonances photoniques. Nous présentons des techniques de nanofabrication avancées pour la création de grands réseaux de trous sur des métaux et de cavités de Fabry-Pérot (FP) nanofluidiques. Ces systèmes sont ensuite utilisés pour étudier, sous régime de couplage fort, les modifications des propriétés de surface et de volume de molécules organiques en phase solide et liquide. En particulier les transitions électroniques de molécules du colorant cyanine en solution liquide ont été couplées à des modes photoniques résonants de cavités FP nanofluidiques spécialement conçues. Leur couplage fort a conduit à une amélioration du rendement quantique d'émission, mettant en évidence la nature radiative des états polaritoniques. / This thesis contributes to the fundamental understanding of the phenomenon of strong coupling of light with organic molecules by implementing new systems and techniques in order to investigate property modifications of molecules coupled with photonic resonances. State-of-the-art nanofabrication techniques for the formation of large hole-array gratings in metals and nanofluidic Fabry-Perot (FP) cavities are presented. These systems were then invested to study, under strong coupling, surface and bulk properties modifications of organic molecules in the solid and liquid phase. In particular, electronic transitions of cyanine dye molecules in liquid solutions were coupled to resonant photonic modes of specially designed nanofluidic FP cavities. Their strong coupling has led to an enhancement of the emission quantum yield, highlighting the radiative nature of the associated polaritonic states.

<p>The fiber-to-the-x (FTTX) has been widely investigated as a leading access technology to meet the ever growing demand for bandwidth in the last mile. The passive optical network (PON) provides a cost-effective and durable solution. In this thesis, we investigate different aspects of the PON, in the search for cost-effective and high-performance designs of link system and devices.</p> <p>In Chapter 2, we propose a novel upstream link scheme for optical orthogonal frequency division multiplexing (OOFDM)-PON. The colorless laser diodes are used at the optical network units (ONUs), and the overlapped channel spectrum of orthogonal subcarrier multiplexing provides high spectral efficiency. At the optical line terminal (OLT), optical switch and all optical fast Fourier transform (OFFT) are adopted for high speed demultiplexing. The deterioration caused by the laser perturbation is also investigated.</p> <p>In Chapter 3, we design a novel polarization beam splitter (PBS), which is one of the most important components in polarization-controlled optical systems, e.g. the next-generation PON utilizing polarization multiplexing. Our PBS is built on a slab waveguide platform where the light is vertically confined. Planar lenses are formed to collimate and refocus light beam by converting the phase front of the beam. A planar subwavelength grating of a wedge shape induces the form birefringence, where the transverse electric (TE) and transverse magnetic (TM) waves have different effective refractive indices, and are steered to distinct directions. This design provides low insertion loss (< 0.9 dB) and low crosstalk (< -30 dB) for a bandwidth of 100 nm in a compact size, and can be realized by different material systems for easy fabrication and/or monolithic integration with other optical components.</p> <p>In Chapter 4, we study the mode partition noise (MPN) characteristics of the Fabry-Perot (FP) laser diode using the time-domain simulation of noise-driven multi-mode laser rate equation. FP laser is cheaper than the widely used distributed feedback (DFB) laser diode in PON, but its MPN is the major limiting factor in an optical transmission system. We calculate the probability density functions for each longitudinal mode. We also investigate the k-factor, which is a simple yet important measure of MPN. The sources of the k-factor are studied with simulation, including the intrinsic source of the laser Langevin noise, and the extrinsic source of the bit pattern.</p> / Doctor of Philosophy (PhD)

Foundation Programme (FP) students at the University of Namibia (UNAM) Oshakati Campus display inadequate academic writing abilities. As their aim is to gain admittance to UNAM main campus science-related courses, it is vital to have effective academic writing skills. This action research (AR) study is a comparison of three writing programmes, the process approach, the modeling approach, and the process genre approach which were implemented separately to three different class groups in 2008 and 2009 to improve the writing skills of students and the teaching practice of the researcher. The effects of the interventions were examined using a combination of the quantitative and qualitative research methods. Data were collected using questionnaires, pre- and post-intervention essays and laboratory reports and interviews. The findings indicate that all three approaches improved the academic writing skills of FP students. The process genre approach had a higher rate of effect than the other two approaches. / English Studies / M. A. (Specialisation TESOL)

Cette thèse s'inscrit dans le contexte de la forte évolution du réseau d'abonnées sur fibre optique (FTTH). Elle a l'objectif d'étudier les solutions d'émetteurs dits "achromatiques" à base de nouveaux composants optoélectroniques pour les réseaux d'accès optiques multiplexés en longueurs d'onde (WDM-PON). Ces émetteurs doivent être peu chers et identiques chez tout abonné quelque soit la longueur d'onde de fonctionnement. Deux types d'émetteurs achromatiques à base de modulateur réflectif sont étudiés dans cette thèse : le laser Fabry-Perot verrouillé par injection optique (IL-FP) et le modulateur électro-absorbant amplifié en réflexion (R-EAM-SOA). Les caractéristiques de l'IL-FP en régime de verrouillage sont évaluées en termes de bruit d'intensité, de modulation, de facteur de couplage phase-amplitude,... La caractérisation tant statique que dynamique est aussi effectuée pour le second composant R-EAM-SOA. D'autres fonctionnalités de ce composant comme la détection directe et la régénération promettent son grand potentiel pour la réalisation d'un dispositif multifonctionnel pour le réseau d'accès. Il est également proposé d'utiliser pour la première fois, le laser à bâtonnets quantiques à blocage de modes (QD-MLL) comme source d'injection multi-longueurs d'onde cohérentes pour l'accès optique WDM. Il est montré qu'il est possible de supprimer le bruit de partition de modes associé à ces structures à blocage de modes. Finalement l'intégration de ces trois composants dans des architectures WDM-PON est réalisée. La faisabilité des systèmes WDM-PON à 2,5 Gb/s à base des IL-FP injectés par une source QD-MLL est évaluée pour différentes configurations et architectures réseau. La montée en débit à 10 Gb/s est prouvée en utilisant le R-EAM-SOA. Un système WDM-PON bidirectionnel avec une capacité de 240 (24 x 10) Gb/s en voie descendante et 60 (24 x 2,5) Gb/s en voie montante a été démontré en utilisant ces solutions d'émetteurs achromatiques proposées. Une des principales sources de dégradation du système WDM-PON bidirectionnel est la rétrodiffusion de Rayleigh pour laquelle le bruit interférométrique produit est étudié. L'impact de ce bruit sur les performances en transmission du système WDM-PON est aussi analysé.

Foundation Programme (FP) students at the University of Namibia (UNAM) Oshakati Campus display inadequate academic writing abilities. As their aim is to gain admittance to UNAM main campus science-related courses, it is vital to have effective academic writing skills. This action research (AR) study is a comparison of three writing programmes, the process approach, the modeling approach, and the process genre approach which were implemented separately to three different class groups in 2008 and 2009 to improve the writing skills of students and the teaching practice of the researcher. The effects of the interventions were examined using a combination of the quantitative and qualitative research methods. Data were collected using questionnaires, pre- and post-intervention essays and laboratory reports and interviews. The findings indicate that all three approaches improved the academic writing skills of FP students. The process genre approach had a higher rate of effect than the other two approaches. / English Studies / M. A. (Specialisation TESOL)