Sensor networks comprise resource-constrained wireless nodes with the capability of gathering information about their surroundings and have recently risen to prominence with the promise of being an effective computing platform for diverse applications, ranging from event detection to environmental monitoring. The database community proposed the use of sensor network query processors (SNQPs) as means to meet data collection requirements using a declarative query language. Declarative queries posed against a sensor network constitute an effective means to repurpose sensor networks and reduce the high software development costs associated with them. The range of sensor network applications is very broad. Such applications have diverse, and often conflicting, QoS expectations in terms of the delivery time of results, the acquisition interval at which data is collected, the total energy consumption of the deployment, or the network lifetime. The conflicting nature of these desiderata is aggravated by the resource-constrained nature of sensor networks as a computing fabric, making it particularly challenging to reconcile the trade-offs that arise. Previously, SNQPs have been focussed on evaluating queries as energy-efficiently as possible. There has been comparatively less work on attempting to meet a broad range of optimization goals and constraints that captured these QoS expectations. In this respect, previous work in SNQP has not aimed at being general purpose across the breadth of applications to which sensor networks have been applied. This PhD dissertation presents an approach for enabling QoS-awareness in SNQPs so that query evaluation plans are generated that exhibit good performance for a broader range of sensor network applications in terms of their QoS expectations. The research contributions reported here include (a) a functional decomposition of the decision-making steps required to compile a declarative query into a query evaluation plan in a sensor network setting; (b) algorithms to implement these decision-making steps; and (c) an empirical evaluation to show the benefits of QoS-awareness compared to a representative fixed-goal SNQP.

Graduation date: 1984

Thesis (Ph. D.)--Virginia Commonwealth University, 2006. / Prepared for: Dept. of Health Administration. Bibliography: leaves 165-173. Also available online.

Thesis (M.Sc. (Agricultural Economics)) --University of Limpopo, 2001.

Data serves as the foundation in building effective deep learning algorithms, yet the process of annotation and curation to maintain high data quality is time-intensive. The challenges arise from the vast diversity and large amount of data, and the inherent complexity in labeling each sample. Then, relying on manual effort to construct high-quality data is implausible and not sustainable in the real world. Instead, this thesis introduces a set of novel techniques to effectively learn from the data with less curation, which is more practical in building AI applications. In this thesis, we systematically study different directions in learning from low-quality data, with a specific focus on visual understanding and being robust to complicated label bias & noise. We first examine the bias exhibited in the whole dataset for image classification, and derive the debiasing algorithms based on representation learning that explores the geometry and distribution of embeddings. In this way, we mitigate the uneven performance over image classes caused by data imbalance, and suppress the spurious correlation between the input images and output predictions such that the model can be better generalized to new classes and maintain robust accuracy with a small number of labeled samples as reference. Then, we extend our analysis to the open-text description of each sample and explore the noisy label in multi-modal pre-training. We build our framework upon contrastive language-image pretraining to learn a common representation space and improve the training effectiveness by automatically eliminating false negative labels and correcting the false positives. Additionally, our approaches show the potential to tackle the label bias in multi-modal training data. Throughout this dissertation, the unifying focus is on the effective approach for learning from low-quality data, which has considered the learning issues from two complementary aspects of data labeling, i.e., the bias in global distribution and the noise in annotation for each sample (local). Different from prior research that are developed on the data with biased & noisy label but artificially simulated from well-curated datasets, our approach has been validated to be resilient to the complex bias and noise in the real-world scenario. We hope our approach can offer contributions to the field of multi-modal machine learning with applications involving real-world low-quality data and the need to avoid manual effort in data construction.

M. Tech. Business Information / Improved Data Quality Management is the desired outcome of managers for many government departments if not all. The constant use of manual systems, the utilization of legacy information systems, the results from departmental mergers and other agency acquisitions, by the South African government departments contributes significantly to poor data quality which leads to fruitless expenditure, fraud and corruption, duplication of processes, none-aligned departmental priorities and service delivery challenges. Literature reveals that there are documents that have been developed to manage the quality of data such as South African Quality Assessment Framework document, which guides and evaluates government organizations on how to achieve quality statistics focusing mainly on the meta-data; however they do not cover other dimensions of Data Quality Management. This research focuses into the strategy of Data Quality Management in government organizations and Data Quality Management tools applied using a conceptual model based on the Technology-Organization-Environment framework.

Least cost formulations for surimi seafood were studied by linear programming (LP) and nonlinear programming (NLP). The effects of water and starches on functional properties of Alaska pollock and Pacific whiting surimi gels were investigated. Six starches (modified potato starch, potato starch, modified wheat starch, wheat starch, modified waxy corn starch, and corn starch) and their mixtures were used as ingredients. Mixture and extreme vertices design were used as experimental designs. Canonical models were applied to the optimization techniques. Blending different kinds of surimi showed linear trends for each functional property, so that LP was successfully employed to optimize surimi lots. Strong interactions were found between surimi and starch or in starch mixtures. Two optimum solutions, obtained from LP and NLP, were compared in this study. Corn starch and modified waxy corn starch greatly improved the functional properties. / Graduation date: 1997

Conselho Nacional de Desenvolvimento Científico e Tecnológico / Potato tubers are used as food and seed. Post-harvest conditions affect tubers for both usages, since physiological changes occur during tuber storage. The objective of this work was to assess the processing quality and time until tuber sprouting of advanced clones of potato. We evaluate thirteen clones and two check varieties of potato in a storage room at 10ºC in two years. The processing quality was evaluated right before loading and after 30, 60, 90 and 120 days of storage for the fresh and dry weight content, chip color and reduced sugars. The experiment was a factorial of clone and storage time in the complete random design, with four repetitions. The percentage of fresh weight loss increased over time during storage, mainly at 120 days. The assessed clones presented dry weight content between 17.4% to 25.2 % during the cold storage. Chip color showed an intense yellow tone and acceptable color classification for the processing industry, even with the increase of reduced sugars during storage. The storage at 10 ºC extended in a 100 days the sprout process. The refrigerated storage promote physiological changes on the tubers, but this changes don´t compromise the use in the industry processing and also allows to increase the storability period of the tubers for both consumption and seeds. / Os tubérculos de batata são destinados tanto para a alimentação quanto para a propagação. Ambas as finalidades são influenciadas pelas condições de armazenamento, que é capaz de promover mudanças fisiológicas nos tubérculos. O objetivo deste trabalho foi avaliar a qualidade de processamento e o tempo de armazenamento até a brotação dos tubérculos de clones avançados de batata. Foram avaliados 13 clones de batata e duas testemunhas, que foram submetidos ao armazenamento em câmera refrigerada a uma temperatura de 10 ºC em dois anos. A qualidade de processamento foi avaliada no momento do armazenamento, imediatamente após a cura, e aos 30, 60, 90 e 120 dias. Os caracteres avaliados foram massa fresca e seca, coloração de chips e o teor de açúcares redutores. Também foi anotado o número de dias para o rompimento da dormência e da dominância apical, quando respectivamente os tubérculos apresentaram um e dois brotos com pelo menos 2 mm de comprimento. O experimento foi um fatorial de clones e épocas de avaliação segundo o delineamento experimental inteiramente casualizado, com quatro repetições. O percentual de perda de massa fresca foi crescente ao longo do armazenamento refrigerado, sobretudo aos 120 dias. Os clones avaliados apresentaram teor de massa seca entre 17,4 % a 25,2 % ao longo do armazenamento refrigerado. A coloração de chips apresentou uma tonalidade amarela intensa e uma classificação aceitável para a indústria, mesmo com o aumento no teor de açúcares redutores após o armazenamento. O armazenamento a 10 ºC prolongou o processo de brotação em até 100 dias de armazenamento. O armazenamento refrigerado promove mudanças fisiológicas nos tubérculos, mas estas mudanças não comprometem a utilização na indústria de processamento, e também possibilita armazenar os tubérculos por um longo período para o consumo e plantio.

Data stream processing in the industrial as well as in the academic field has gained more and more importance during the last years. Consider the monitoring of industrial processes as an example. There, sensors are mounted to gather lots of data within a short time range. Storing and post-processing these data may occasionally be useless or even impossible. On the one hand, only a small part of the monitored data is relevant. To efficiently use the storage capacity, only a preselection of the data should be considered. On the other hand, it may occur that the volume of incoming data is generally too high to be stored in time or–in other words–the technical efforts for storing the data in time would be out of scale. Processing data streams in the context of this thesis means to apply database operations to the stream in an on-the-fly manner (without explicitly storing the data). The challenges for this task lie in the limited amount of resources while data streams are potentially infinite. Furthermore, data stream processing must be fast and the results have to be disseminated as soon as possible. This thesis focuses on the latter issue. The goal is to provide a so-called Quality-of-Service (QoS) for the data stream processing task. Therefore, adequate QoS metrics like maximum output delay or minimum result data rate are defined. Thereafter, a cost model for obtaining the required processing resources from the specified QoS is presented. On that basis, the stream processing operations are scheduled. Depending on the required QoS and on the available resources, the weight can be shifted among the individual resources and QoS metrics, respectively. Calculating and scheduling resources requires a lot of expert knowledge regarding the characteristics of the stream operations and regarding the incoming data streams. Often, this knowledge is based on experience and thus, a revision of the resource calculation and reservation becomes necessary from time to time. This leads to occasional interruptions of the continuous data stream processing, of the delivery of the result, and thus, of the negotiated Quality-of-Service. The proposed robustness concept supports the user and facilitates a decrease in the number of interruptions by providing more resources.

Potato (Solanum tuberosum L.) is the most important horticultural crop of the world. Late blight (Phytophthora infestans (Mont.) de Bary) is the most important and devastating disease. There is no cultivar with sufficient level of resistance in Brazil being most part of production costs used for its control. Besides late blight resistance, new cultivars should have high tuber quality either for processing or table stock. The objectives of this study were to a) develop progeny of late blight resistant progenitors; b) compare selection for tuber quality in single and eight hill plots; c) identify clones with high tuber quality for processing or table stock; and d) study culture medium to increase growth and sporulation of P. infestans isolates. rosses among late blight resistant clones (SMIJ319-1, SMIJ456-4Y, SMIG274-3, Torridon, Tollocan e Stirling) and clones or cultivars either with high tuber quality or highly adapted were made in a screen house of Fitotecnia Department of Universidade Federal de Santa Maria during 2005. Twenty families were evaluated in single hill during spring 2006. Selected clones were also evaluated in eight hill plots during autumn 2006. Tuber appearance, specific gravity and chip color were assessed in all selected clones of both clonal generations. The 19 selected new clones were then evaluated in twenty hill plots during spring 2006 for tuber appearance, chip color, contents of dry matter, reduced sugars, starch and amylose. Growth and sporulation of P. infestans isolates were assessed in potato-dextrose-agar (PDA), carrot-agar, corn flour-agar, rye B and juice of eight vegetables (V8) culture mediums. A high percentage of clones combined tuber appearance, specific gravity and chip color and were identify in progeny of Tollocan, Stirling, SMIJ319-1 e SMIJ456-4Y. A moderate selection index resulted in 89.4%, 63.9% and 79.8% selections in single hill and 71.2%, 63.9% and 71.4% selections in eight hill plots respectively for tuber appearance, specific gravity and chip color. The clones SMA501-1, SMA508-2, SMA508-4, SMA514-8, SMA514-10, SMA514-11 e SMA520-5 showed high processing potential. The clones SMA504-2, SMA505-2, SMA505-7, SMA506-4, SMA513-2, SMA516-2, SMA517-2, SMA519-1 e SMA520-5 showed high potential either for processing or table stock. The rye B medium increased mycelium growth of P. infestans and the culture mediums PDA and V8 increased sporangium production / A batata (Solanum tuberosum L.) é a olerícola cultivada de maior expressão econômica no mundo e no Brasil,. A principal e mais devastadora doença da cultura é a requeima, causada pelo oomiceto Phytophthora infestans (Mont.) de Bary e a ausência de cultivares brasileiras, com altos níveis de resistência determina que uma grande parcela dos custos de produção sejam empregados para o seu controle. A resistência a requeima por si não determina a aceitação de novas cultivares, que devem apresentar alta qualidade de tubérculo para processamento ou para consumo de mesa. Os objetivos deste estudo foram a) desenvolver progênies de genitores resistentes a requeima; b) comparar a seleção para qualidade de tubérculo em gerações de cova única e oito covas; c) identificar clones de alta qualidade de tubérculo com potencial para processamento industrial ou consumo; e d) determinar o meio de cultura que proporcione o maior crescimento e esporulação de isolados de P. infestans. Foram realizados cruzamentos em telado do Departamento de Fitotecnia da Universidade Federal de Santa Maria durante o ano de 2005. Clones resistentes a requeima (SMIJ319-1, SMIJ456- 4Y, SMIG274-3, Torridon, Tollocan e Stirling) foram cruzados com clones ou cultivares adaptados ou de alta qualidade de tubérculo, resultando em 20 famílias que foram avaliadas e clones selecionados na geração de cova única na primavera de 2005. Os clones selecionados foram também avaliados na geração de oito covas no outono de 2006. A seleção para aparência de tubérculo, gravidade específica e coloração de chips resultou em 19 clones que foram novamente plantados em parcelas de 20 covas na primavera de 2006 e avaliados para aparência de tubérculo, coloração de chips, teores de matéria seca, açúcares redutores, amido e amilose. O crescimento e a esporulação de isolados de P. infestans foram avaliados nos meios batata-dextrose-ágar (BDA), cenoura-ágar, farinha de milho-ágar, B de centeio e suco de oito vegetais (V8). Alta porcentagem de clones que combinam aparência de tubérculo, gravidade específica e coloração de chips foram identificados nas progênies de Tollocan, Stirling, SMIJ319-1 e SMIJ456-4Y. A aplicação de moderado índice de seleção permitiu a identificação de 89%, 63,9% e 79,8% dos clones em cova única e 71,2%, 63,9% e 71,4% em oito covas respectivamente para aparência de tubérculo, gravidade específica e coloração de chips. Os clones SMA501-1, SMA508-2, SMA508-4, SMA514-8, SMA514-10, SMA514-11 e SMA520-5 apresentaram alto potencial para processamento industrial e os clones SMA504-2, SMA505-2, SMA505-7, SMA506-4, SMA513-2, SMA516-2, SMA517-2, SMA519-1 e SMA520-5 apresentam potencial para consumo de mesa ou processamento na forma de fritas tipo palito. O meio B de centeio promoveu o rápido crescimento micelial de P. infestans e os meios de cultura BDA e suco V8 promoveram a maior produção de esporângios