Over recent decades, environmental regulations on wastewater treatment plants (WWTP) have trended towards increasingly stringent nutrient removal requirements for the protection of local waterways. However, such regulations ignore the other environmental impacts that might accompany the apparent improvements to the WWTP. This PhD thesis used Life Cycle Assessment (LCA) to quantify these environmental trade-offs, and so better inform policy makers on the wider benefits and burdens associated with wastewater treatment. A particular focus was also given to the generation of methane and nitrous oxide in wastewater systems, since the quantification of greenhouse gas (GHG) emissions from WWTPs is presently a substantial area of uncertainty. Rapid changes to the GHG regulatory landscape mean that this level of uncertainty, now represents an unacceptable business risk for many water utilities. Specifically, there were three research objectives of this thesis: Research Objective No.1 – Environmental optimisation of wastewater treatment systems – For typical receiving environments, the optimum wastewater treatment system configuration is not necessarily at the limit of best practice for nutrient removal. The LCA approach to this research objective was divided into two stages. In stage I, a comprehensive desk-top life cycle inventory of ten different wastewater treatment scenarios was completed. The scenarios covered six process configurations and treatment standards ranging from raw sewage to advanced nutrient removal. It was shown that physical infrastructure, chemical usage and operational energy all increased with the level of nutrient removal. These trends represented a trade-off of negative environmental impacts against improved local receiving water quality. In stage II of the LCA, a quantitative life cycle impact assessment of the ten scenarios, referenced against Australian normalisation data, was completed. From a normalised perspective against Australian society, the contribution of WWTPs to headline issues such as global warming and energy consumption was found to be very small. The more prominent environmental impact categories were eutrophication due to nutrient discharge and toxicity issues, due to heavy metals in biosolids. There existed a broader environmental trade-off for nutrient removal, that could only be justified by society and regulators implicitly placing higher value on local water quality, than on other global environmental pressures. In light of this quantitative LCA, regulatory agencies should consider the broader environmental consequences of their policies such as the Queensland Water Quality Guidelines. It is suggested that the scope of WWTP licensing considerations should be widened from a singular focus on water quality objectives, to a more comprehensive LCA-based approach. Research Objective No. 2 – Quantification of nitrous oxide emissions from biological nutrient removal (BNR) wastewater treatment plants – Current GHG assessment methods for wastewater treatment plants are grossly inaccurate because of significant unaccounted N2O emissions. The research for objectives two and three was funded by the Water Services Association of Australia (WSAA), which is the peak body of the Australian urban water industry. Thus, whilst the earlier LCA results suggested that GHG emissions from WWTPs were insignificant from a national perspective, the industry is actually very engaged on this issue from an environmental responsibility and business risk perspective. This PhD study adopted a rigorous mass balance approach to determine N2O-N generation at seven full-scale WWTPs. The results varied considerably in the range 0.006 – 0.253 kgN2O-N generated per kgNdenitrified (average: 0.035 +/- 0.027). These results were generally larger than the current default value assumed in the National Greenhouse and Energy Reporting (Measurement) Technical Guidelines (i.e. 0.01 kg N2O-N.kgN-1denitrified). High N2O-N generation was shown to correspond with elevated bulk NO2--N concentrations in the bioreactor. The results also suggested that WWTPs designed for low effluent TN have lower and less variable N2O generation than plants that only achieve partial denitrification. Research Objective No.3 – Quantification of methane emissions from low-strength wastewater collection systems – Current default GHG assessment methods for sewerage systems are grossly inaccurate because of significant unaccounted CH4 emissions from rising mains. Presently, international GHG guidelines state that “wastewater in closed underground sewers is not believed to be a significant source of methane” (IPCC, 2006). However, the results of this PhD research demonstrated that methane generation in rising main sewers is substantial. It was shown that dissolved methane concentrations were dependent upon pipeline geometry and sewage residence time. Consequently, it was possible to develop a simple, yet robust, theoretical model that predicted methane generation from these two independent parameters. This model provides a practical means for water authorities globally to make an estimate of the currently unaccounted methane emissions from pressurised sewerage systems.

An essential step to achieving human-machine speech communication with the naturalness of communication between humans is developing a machine that is capable of recognising emotions based on speech. This thesis presents research addressing this problem, by making use of acoustic and prosodic information. At a feature level, novel group delay and weighted frequency features are proposed. The group delay features are shown to emphasise information pertaining to formant bandwidths and are shown to be indicative of emotions. The weighted frequency feature, based on the recently introduced empirical mode decomposition, is proposed as a compact representation of the spectral energy distribution and is shown to outperform other estimates of energy distribution. Feature level comparisons suggest that detailed spectral measures are very indicative of emotions while exhibiting greater speaker specificity. Moreover, it is shown that all features are characteristic of the speaker and require some of sort of normalisation prior to use in a multi-speaker situation. A novel technique for normalising speaker-specific variability in features is proposed, which leads to significant improvements in the performances of systems trained and tested on data from different speakers. This technique is also used to investigate the amount of speaker-specific variability in different features. A preliminary study of phonetic variability suggests that phoneme specific traits are not modelled by the emotion models and that speaker variability is a more significant problem in the investigated setup. Finally, a novel approach to emotion modelling that takes into account temporal variations of speech parameters is analysed. An explicit model of the glottal spectrum is incorporated into the framework of the traditional source-filter model, and the parameters of this combined model are used to characterise speech signals. An automatic emotion recognition system that takes into account the shape of the contours of these parameters as they vary with time is shown to outperform a system that models only the parameter distributions. The novel approach is also empirically shown to be on par with human emotion classification performance.

Thesis (MScEng) -- Stellenbosch University, 2014. / ENGLISH ABSTRACT: The increasing popularity of microblogging services such as Twitter means that more and more unstructured data is available for analysis. The informal language usage in these media presents a problem for traditional text mining and natural language processing tools. We develop a pre-processor to normalise this noisy text so that useful information can be extracted with standard tools. A system consisting of a tokeniser, out-of-vocabulary token identifier, correct candidate generator, and N-gram language model is proposed. We compare the performance of generative and discriminative probabilistic models for these different modules. The effect of normalising the training and testing data on the performance of a tweet sentiment classifier is investigated. A linear-chain conditional random field, which is a discriminative model, is found to work better than its generative counterpart for the tokenisation module, achieving a 0.76% character error rate compared to 1.41% for the finite state automaton. For the candidate generation module, however, the generative weighted finite state transducer works better, getting the correct clean version of a word right 36% of the time on the first guess, while the discriminatively trained hidden alignment conditional random field only achieves 6%. The use of a normaliser as a pre-processing step does not significantly affect the performance of the sentiment classifier. / AFRIKAANSE OPSOMMING: Mikro-webjoernale soos Twitter word al hoe meer gewild, en die hoeveelheid ongestruktureerde data wat beskikbaar is vir analise groei daarom soos nooit tevore nie. Die informele taalgebruik in hierdie media maak dit egter moeilik om tradisionele tegnieke en bestaande dataverwerkingsgereedskap toe te pas. ’n Stelsel wat hierdie ruiserige teks normaliseer word ontwikkel sodat bestaande pakkette gebruik kan word om die teks verder te verwerk. Die stelsel bestaan uit ’n module wat die teks in woordeenhede opdeel, ’n module wat woorde identifiseer wat gekorrigeer moet word, ’n module wat dan kandidaat korreksies voorstel, en ’n module wat ’n taalmodel toepas om die mees waarskynlike skoon teks te vind. Die verrigting van diskriminatiewe en generatiewe modelle vir ’n paar van hierdie modules word vergelyk en die invloed wat so ’n normaliseerder op die akkuraatheid van ’n sentimentklassifiseerder het word ondersoek. Ons bevind dat ’n lineêre-ketting voorwaardelike toevalsveld—’n diskriminatiewe model — beter werk as sy generatiewe eweknie vir tekssegmentering. Die voorwaardelike toevalsveld-model behaal ’n karakterfoutkoers van 0.76%, terwyl die toestandsmasjien-model 1.41% behaal. Die toestantsmasjien-model werk weer beter om kandidaat woorde te genereer as die verskuilde belyningsmodel wat ons geïmplementeer het. Die toestandsmasjien kry 36% van die tyd die regte weergawe van ’n woord met die eerste raaiskoot, terwyl die diskriminatiewe model dit slegs 6% van die tyd kan doen. Laastens het ons bevind dat die vooraf normalisering van Twitter boodskappe nie ’n beduidende effek op die akkuraatheid van ’n sentiment klassifiseerder het nie.

La normalisation occupe une place primordiale dans le domaine des technologies de l’information et de la communication, ainsi que pour les TICE, en permettant d’assurer une interopérabilité entre les différents systèmes et équipements. Dans ce contexte rentre notre travail de recherche qui vise à reconsidérer la perspective de mise en place des systèmes de gestion de corpus lexicaux homogènes, multilingues, partageables et interopérables, ainsi que la mise en valeur indispensable de la normalisation pour l’éducation et ses retombées bénéfiques en termes d’accessibilité, d’interopérabilité et d’adaptabilité…Une étude expérimentale qui vise à tester certains outils informatiques qui servent à la production et à la gestion d’une terminologie multilingue et normalisée constitue la facette expérimentale de notre travail. L’observation, la comparaison, la découverte et l’analyse du comportement des usagers est faîte à partir d’une enquête. / Standardization holds a pivotal place in the domain of information and communicationtechnologies as well as for the TICE as it allows for a better harmony between the differentsystems and equipments. Our research project is situated in this context. It aims to reconsiderthe perspective of elaborating systems of management of homogenous, multilingual,shareable and interoperable lexical corpora. It also emphasizes the necessity ofstandardization and its effects on education in terms of accessibility, interoperability andadaptability…An experimental study which aims at testing some computer tools used in theproduction and management of a standardized and multilingual terminology constitutes thecorpus of our work. The observation, comparison, discovery and analysis of users behaviouris based on a survey. / تحظى المعاییر الدولیة بمكانة ھامة في مجال استعمال تكنولوجیا المعلومات والاتصال ولسیما المتصلة بالتعلیم والتربیة والتكوین. ذلك.ان عملیة التعییر تضمن سلامة التواصل بین مختلف مكونات نظم وأدوات التعلیمفي ھذا الإطار یندرج بحثنا الذي یھدف الى اعادة النظر في نظم انتاج ومعالجة المدونات المعجمیة بطریقة تجعل منھا مدوناتمتجانسة ومتعددة اللغات وقابلة للاستعمال المتبادل. كما نبین من خلال ھذا البحث الحاجة الى وضع جملة من المعاییر الدولیة في.میدان التعلیم تخص طریقة الوصول الى المعلومة والقدرة على التكیف مع النظم التعلیمیةھذا البحث عبارة عن دراسة تجریبیة لبعض البرمجیات الاعلامیة لقصد جعلھا قابلة لإنتاج ومعالجة المصطلحات متعددة اللغات.مطابقة في ذلك المعاییر الدولیة

Cette thèse porte sur les espaces publics à Johannesburg, capitale économique de l’Afrique du Sud. Dans le contexte contemporain, l’utilisation de la notion occidentale d’espaces publics pose problème : d’une part, du fait des ségrégations passées qui ont eu tendance à faire de ces espaces des lieux de séparation et de mise à distance des différents publics ; et d’autre part, du fait des forts taux de violence et du fort sentiment d’insécurité, qui tendent à encourager la sécurisation et la privatisation de ces mêmes espaces. L’enjeu est alors de comprendre les éventuels processus de construction de la publicité (au sens de caractère public) de ces espaces, à la fois sur le plan juridique, social et politique. Pour ce faire, l’art qui se déploie dans les espaces juridiquement publics de la métropole depuis la fin de l’apartheid, est utilisé comme une clef de lecture privilégiée de ces phénomènes, en tant qu’il permettrait, ainsi que nous entendons le montrer, de créer des espaces de rencontre et de débats ou, à l’inverse, de mieux réguler et contrôler ces espaces. Selon une approche qualitative, notre étude se base à la fois sur des observations de terrain et sur des entretiens conduits auprès des producteurs mais aussi des récepteurs de cet art qui a lieu dans les espaces publics. A la croisée de la géographie urbaine et de la géographie culturelle, nous envisageons donc de réexaminer la notion d’espaces publics au prisme de l’art à Johannesburg en vue de saisir – entre tentative de normalisation et résistance à cette normalisation – quelle ville est aujourd’hui à l’œuvre non seulement à Johannesburg, mais aussi, à travers elle, dans d’autres villes du monde. / This Ph.D. thesis deals with public spaces in Johannesburg, the economic capital of South Africa. In the current context, the issues raised by the use of the western notion of public spaces are explored. On one hand, the previous segregations tended to mark off spaces into different publics completely separated from each other. On the other hand, the high rates of violence and sense of insecurity enhance securitization and privatization of these same spaces. What is at stake is to understand how the publicness of these spaces can be legally, socially, and politically built. In that framework, art spread in legally public spaces of Johannesburg since the end of apartheid is used as a tool to understand and reveal these phenomena since it is presented, as we aim at demonstrating, as a mean to create spaces of encounter and debate or, conversely, to regulate and control better these spaces. In a qualitative approach, our study is based on field observations and interviews with both producers and receivers of this art which takes place in public spaces. At the crossroads of urban geography and cultural geography, we are therefore re-examining the concept of public spaces through the prism of art in Johannesburg to figure out – between normalization and resistance to this normalization – which city is today at work not only in Johannesburg, but also, through her, in other cities of the world.

L’éducation du patient adulte entre normalisation et normativité. Le développement de l’éducation du patient tente de répondre à l’inexorable progression du diabète et des autres maladies chroniques. L’objectif de cette thèse est d’interroger les rapports aux normes de santé des malades et des soignants. Il s’agit de comprendre ce qui sous-tend le développement de cette pratique avec une médecine qui progresse dans son approche scientifique. La pédagogie devient alors prescription à trouver la norme de santé. La substitution d’une norme médicale par une norme pédagogique pose question sur la démarche éducative entre normalisation et normativité. La normalisation expose le patient aux normes issues d’une médecine fondée sur les sciences alors que la normativité invite le malade à la recherche de ses propres normes de vie avec sa maladie. La démarche éthique permet de questionner l’autonomie prescrite et nous invite à partager nos impuissances pour mieux partager nos savoirs. Le courage nous paraît au cœur de la délibération pour adapter la démarche éducative à chaque situation singulière et permettre à l’éducation du patient d’être thérapeutique. / Education of adult patients : between normalization and normativity. The development of patient education attempts to address the relentless progression of diabetes and other chronic diseases. The objective of this thesis is to examine therelationships between medical standards for patients and caregivers. It aims to understand what underlies the development of this practice with medicine that is progressing in its scientific approach. Then pedagogy becomes a prescription to find health standards. There placement of a medical standard with an educational standard brings into question the educational process between normalization and normativity. Normalization exposes thepatient to the rules that come from science-based medicine while normativity invites the patient to find their own standards in order to live with their illness. The ethical approach makes it possible to question the prescribed autonomy and invites us to share ourweaknesses in order to better share our knowledge. Courage seems to be at the core of the deliberation to adapt an educational approach to each individual situation to enable patient education to be therapeutic.

Oral health promotion interventions (OHPI) seek to achieve sustainable oral health improvements through actions directed at the underlying determinants of oral health. Clinical trials often promise levels of effectiveness which are not actually achieved in general use, indicating problems with implementation. This study set out to identify an OHPI for which there was strong evidence of effectiveness and then aimed to explore issues that arose in the implementation of that intervention. Research methods A sequential portfolio design was utilised. The first phase involved systematic reviews of literature to identify existing community based OHPI and interventions shown to be effective in reducing dental caries. These were also examined to determine whether the existing evidence base was informative about the process of implementing best evidence. Systematic reviews were undertaken on effectiveness of: supervised toothbrushing with fluoridated toothpastes, supervised fluoridated mouthrinsing and promotion of dietary behaviours, all, in school children. MEDLINE, CINAHL, Cochrane Central Register of Controlled Trials, EMBASE and BNI electronic databases were searched for articles published between 1990 and 2013. In total, 47 studies were included in the reviews: 21 studies on supervised toothbrushing with fluoridated toothpastes, 12 on fluoridated mouthrinsing and 14 studies on healthy dietary behaviours towards caries prevention. The second phase was a qualitative research study, which was undertaken to explore the experiences, and perceptions of participants involved in decision-making, planning and delivery of one specific community based OHPI (supervised toothbrushing) in the North East of England. Nineteen participants (NHS and school staff) participated in the qualitative study. Knowledge translation in oral health promotion One-to-one interviews and focus groups interviews were conducted with the participants. Normalisation Process Theory constructs informed the development of the interview guides and was also used in data analysis. In the third phase of the study, an integration of the findings of the first and second phases of the study was conducted. Results In the systematic reviews of literature conducted, the various levels of clinical effectiveness identified were influenced by appropriate and regular fluoride use, involvement of OHP specialists, supervision by parents and the free provision and availability of materials. There was however a general lack of comprehensive information on all aspects of implementation of OHPI. In the second phase of the study, an exploration of barriers to implementation of an OHPI revealed the following: -inadequate utilisation of research evidence -gaps in leadership and management structures -non- investment in engagement and ownership of the intervention -challenges with partnership working and evaluation of implementation of OHPI. Conclusions Availability of evidence of effectiveness of an intervention does not imply that the intervention will be successful when rolled out. Published articles on evidence-based interventions do not have comprehensive information on the mechanisms and workability of the processes required for effective implementation of OHPI. Normalisation Process Theory (NPT) enabled the exploration of factors that could facilitate knowledge translation and successful implementation of OHPI. In previous studies, NPT was used to evaluate effectiveness of interventions; in this study it was used to explore the implementation process of an OHPI and has highlighted the need for Knowledge translation in oral health promotion oral health promotion strategy makers and commissioners to revisit the “sense-making” aspect of evidence implementation, to reflect on the need for investing in all members of the team, to encourage the ‘ownership’ of interventions being implemented. In addition, there is a need to review existing leadership and management structures and to re-examine and amend the processes by which OHPI are monitored and reported. These measures would enable maximised effectiveness and sustainability of clinically effective OHPI.

Cette thèse de doctorat interroge la polysémie des modalités de professionnalisation du fait professionnel contemporain. Elle s’inscrit au croisement de la sociologie du travail et des groupes professionnels se centrant sur les aides familiales comme clef d’entrée d’une analyse de l’évolution de l’autonomie professionnelle. Les modalités de professionnalisation au sens large restant l’objet substantif. Nous abordons la professionnalisation comme un folk concept afin de saisir les enjeux centraux au cœur de ces processus, que sont la division du travail et l’adéquation formation-emploi. Ce faisant, nous traduisons la professionnalisation comme un processus temporel mais aussi social. Ce mouvement dialectique s’initie de façon endogène par un groupe de travailleurs et est négocié de façon exogène par des acteurs extérieurs entrant en relation avec ces travailleurs. En ce sens, nous considérons que la professionnalisation est un prisme pour examiner les rapports sociaux de travail qui cristallisent le fait professionnel contemporain.<p><p>Nous avons choisi de centrer notre analyse sur un groupe professionnel méconnu, en recherche de reconnaissance et dont les incertitudes sur le travail pèsent à cause des redéfinitions émises par les politiques d’emplois le concernant suite aux changements démographiques et sociétaux affectant l’ensemble de la société. Notre hypothèse centrale est que la professionnalisation désignent des phénomènes problématiques ;qu’elle ne vise pas exclusivement à autonomiser un groupe de travailleurs sur le marché du travail afin qu’ils obtiennent un statut. Elle peut également servir à valoriser une activité afin de légitimer et développer la reconnaissance des organisations prestataires du dit travail.<p><p>Pour ce faire, nous sommes partis d’un cadre analytique original et novateur croisant les différentes modalités mentionnées dans la littérature pour en faire une typologie permettant de saisir l’hétérogénéité des phénomènes que recouvre la professionnalisation. Nous avons commencé par retracer les conditions socio-historiques de constitution du groupe à travers une recherche documentaire agrémentée par un travail ethnographique et d’observations participantes. Afin de tester nos hypothèses, nous avons opté pour des méthodologies mixtes. 412 questionnaires furent ainsi récoltés auprès des fédérations employeuses du secteur pour cerner les identités professionnelles revendiquées et convoquées. De même, une trentaine d’entretiens semi-dirigés ont été réalisé pour approfondir l’analyse en termes de trajectoires biographiques et d’évolutions des pratiques.<p><p>Nos principaux résultats se traduisent par une innovation théorique, proposant de replacer des formes d’analyses structurannionistes au sein de la sociologie des groupes professionnelles. Ce faisant, nous constatons que le processus de professionnalisation des aides familiales a suivi une catégorisation d’intervention publique où professionnaliser était synonyme de création d’emploi pour in fine devenir un instrument idéologique, qui est repris progressivement sous une catégorisation gestionnaire détachant l’individu, pris comme travailleur professionnel, de la fonction qu’il exerce.<p> / Doctorat en Sciences politiques et sociales / info:eu-repo/semantics/nonPublished

L'épigénomique pourrait nous aider à mieux comprendre pourquoi différents types cellulaires montrent différents comportements. Puisque, dans le cadre d'études épigénétiques, il peut êtrenécessaire de comparer plusieurs profils de séquençage, il y a un besoin urgent en nouvelles approches et nouveaux outils pour pallier aux variabilités techniques sous-jacentes. Nous avons développé NGS-QC, un système de contrôle qualité qui détermine la qualité de données et Epimetheus, un outil de normalisation d'expériences de modifications d'histones basé sur les quartiles afin de corriger les variations techniques entre les expériences. Enfin, nous avons intégré ces outils dans un pipeline d'analyse allèle-spécifique afin de comprendre le statut épigénétique de XCI dans le cancer du sein où la perte du Xi est fréquent. Notre analyse a dévoilé des perturbations dans le paysage épigénétique du X et des réactivations géniques aberrantes dans le Xi, dont celles associées au développement du cancer. / Epigenomics would help us understand why various cells types exhibit different behaviours. Aberrant changes in reversible epigenetic modifications observed in cancer raised focus towards epigenetic targeted therapy. As epigenetic studies may involve comparing multi-profile sequencing data, thereis an imminent need for novel approaches and tools to address underlying technical variabilities. Wehave developed NGS-QC, a QC system to infer the experimental quality of the data and Epimetheus, a quantile-based multi-profile normalization tool for histone modification datasets to correct technical variation among samples. Further, we have employed these developed tools in an allele-specific analysis to understand the epigenetic status of X chromosome inactivation in breast cancer cells where disappearance of Xi is frequent. Our analysis has revealed perturbation in epigenetic landscape of X and aberrant gene reactivation in Xi including the ones that are associated with cancer promotion.

La sclérose en plaques (SEP) est une maladie auto-immune inflammatoire du jeune adulte causant des handicaps variables et progressifs irréversibles. Cette maladie est présente de manière prépondérante dans l’hémisphère nord. Cette thèse s’attache à la caractérisation et à la modélisation de signatures IRM multimodales des lésions de sclérose en plaques. L’objectif est d’améliorer les modèles de représentation de l’image et d’adapter les méthodes d’apprentissage pour la reconnaissance visuelle, dans le cas où des informations de haut niveau telles que les lésions SEP incluses dans l’IRM sont extraites. Nous proposons dans cette thèse un nouvel algorithme de normalisation d’intensité en IRM, particulièrement centré sur la normalisation d’images longitudinales multimodales, afin de produire des détections d’évolution de lésion robustes. Cette normalisation est centrée sur la modélisation de l’histogramme de l’image par un modèle de mixture de Gaussiennes robuste à la présence de lésions. Faisant suite à cet algorithme, nous proposons également deux nouvelles méthodes de détection de lésions SEP basées sur (1) une comparaison statistique du patient vis à vis d’une population de sujets contrôle et (2) un cadre probabiliste de détection basé sur un apprentissage d’une classe (tissus sains). Nous avons évalué les algorithmes proposés sur plusieurs jeux de données multi-centriques et vérifié leur efficacité dans la détection de lésions. / Multiple Sclerosis (MS) is an acquired inflammatory disease, which causes disabilities in young adults and it is common in northern hemisphere. This PhD work focuses on characterization and modeling of multidimensional MRI signatures in MS Lesions (MSL). The objective is to improve image representation and learning for visual recognition, where high level information such as MSL contained in MRI are automatically extracted. We propose a new longitudinal intensity normalization algorithm for multichannel MRI in the presence of MS lesions, which provides consistent and reliable longitudinal detections. This is primarily based on learning the tissue intensities from multichannel MRI using robust Gaussian Mixture Modeling. Further, we proposed two MSL detection methods based on a statistical patient to population comparison framework and probabilistic one class learning. We evaluated our proposed algorithms on multi-center databases to verify its efficacy.