Placement and compaction of the backfill behind an earth retaining wall may induce a vertical shear force at the soil-to-wall interface. This vertical shear force, or downdrag, is beneficial for the stability of the structure. A significant reduction in construction costs may result if the downdrag is accounted for during design. This potential reduction in costs is particularly interesting in the case of U.S. Army Corps of Engineers lock walls. A simplified procedure is available in the literature for estimating the downdrag force developed at the wall-backfill interface during backfilling of a retaining wall. However, finite element analyses of typical U.S. Army Corps of Engineers lock walls have shown that the magnitude of the downdrag force may decrease during operation of the lock with a rise in the water table in the backfill. They have also shown that pre- and post-construction stress paths followed by interface elements often involve simultaneous changes in shear and normal stresses and unloading-reloading. The hyperbolic formulation for interfaces (Clough and Duncan 1971) is accurate for modeling the interface response in the primary loading stage under constant normal stress. However, it has not been extended to model simultaneous changes in shear and normal stresses or unloading-reloading of the interface. The purpose of this research was to develop an interface model capable of giving accurate predictions of the interface response under field loading conditions, and to implement this model in a finite element program. In order to develop the necessary experimental data, a series of tests were performed on interfaces between concrete and two different types of sand. The tests included initial loading, staged shear, unloading-reloading, and shearing along complex stress paths. An extended hyperbolic model for interfaces was developed based on the results of the tests. The model is based on Clough and Duncan (1971) hyperbolic formulation, which has been extended to model the interface response to a variety of stress paths. Comparisons between model calculations and tests results showed that the model provides accurate estimates of the response of interfaces along complex stress paths. The extended hyperbolic model was implemented in the finite element program SOILSTRUCT-ALPHA, used by the U.S. Army Corps of Engineers for analyses of lock walls. A pilot-scale test was performed in the Instrumented Retaining Wall (IRW) at Virginia Tech that simulated construction and operation of a lock wall. SOILSTRUCT-ALPHA analyses of the IRW provided accurate estimates of the downdrag magnitude throughout inundation of the backfill. It is concluded that the extended hyperbolic model as implemented in SOILSTRUCT-ALPHA is adequate for routine analyses of lock walls. / Ph. D.

Currently, one of the most common causes of death worldwide is cancer. The development of innovative methods to support the early and accurate detection of cancers is required to increase the recovery rate of patients. Several studies have shown that medical Hyperspectral Imaging (HSI) combined with artificial intelligence algorithms is a powerful tool for cancer detection. Various preprocessing methods are commonly applied to hyperspectral data to improve the performance of the algorithms. However, there is currently no standard for these methods, and no studies have compared them so far in the medical field. In this work, we evaluated different combinations of preprocessing steps, including spatial and spectral smoothing, Min-Max scaling, Standard Normal Variate normalization, and a median spatial smoothing technique, with the goal of improving tumor detection in three different HSI databases concerning colorectal, esophagogastric, and brain cancers. Two machine learning and deep learning models were used to perform the pixel-wise classification. The results showed that the choice of preprocessing method affects the performance of tumor identification. The method that showed slightly better results with respect to identifing colorectal tumors was Median Filter preprocessing (0.94 of area under the curve). On the other hand, esophagogastric and brain tumors were more accurately identified using Min-Max scaling preprocessing (0.93 and 0.92 of area under the curve, respectively). However, it is observed that the Median Filter method smooths sharp spectral features, resulting in high variability in the classification performance. Therefore, based on these results, obtained with different databases acquired by different HSI instrumentation, the most relevant preprocessing technique identified in this work is Min-Max scaling.

Utilizing multiple biomarkers in medical research is crucial for the diagnostic accuracy of detecting diseases. An optimal method for combining these biomarkers is essential to maximize the Area Under the Receiver Operating Characteristic (ROC) Curve (AUC). The optimality of the likelihood ratio has been proven but the challenges persist in estimating the likelihood ratio, primarily on the estimation of multivariate density functions. In this study, we propose a non-parametric approach for estimating multivariate density functions by utilizing Smoothing Spline density estimation to approximate the full likelihood function for both diseased and non-diseased groups, which compose the likelihood ratio. Simulation results demonstrate the efficiency of our method compared to other biomarker combination techniques under various settings for generated biomarker values. Additionally, we apply the proposed method to a real-world study aimed at detecting childhood autism spectrum disorder (ASD), showcasing its practical relevance and potential for future applications in medical research. Change point detection for functional time series has attracted considerable attention from researchers. Existing methods either rely on FPCA, which may perform poorly with complex data, or use bootstrap approaches in forms that fall short in effectively detecting diverse change functions. In our study, we propose a novel self-normalized test for functional time series implemented via a non-overlapping block bootstrap to circumvent reliance on FPCA. The SN factor ensures both monotonic power and adaptability for detecting diverse change functions on complex data. We also demonstrate our test's robustness in detecting changes in the autocovariance operator. Simulation studies confirm the superior performance of our test across various settings, and real-world applications further illustrate its practical utility. / Doctor of Philosophy / In medical research, it is crucial to accurately detect diseases and predict patient outcomes using multiple health indicators, also known as biomarkers. Combining these biomarkers effectively can significantly improve our ability to diagnose and treat various health conditions. However, finding the best way to combine these biomarkers has been a long-standing challenge. In this study, we propose a new, easy-to-understand method for combining multiple biomarkers using advanced estimation techniques. Our method takes into account various factors and provides a more accurate way to evaluate the combined information from different biomarkers. Through simulations, we demonstrated that our method performs better than other existing methods under a variety of scenarios. Furthermore, we applied our new method to a real-world study focusing on detecting childhood autism spectrum disorder (ASD), highlighting its practical value and potential for future applications in medical research. Detecting changes in patterns over time, especially shifts in averages, has become an important focus in data analysis. Existing methods often rely on techniques that may not perform well with more complex data or are limited in the types of changes they can detect. In this study, we introduce a new approach that improves the accuracy of detecting changes in complex data patterns. Our method is flexible and can identify changes in both the mean and variation of the data over time. Through simulations, we demonstrate that this approach is more accurate than current methods. Furthermore, we applied our method to real-world climate research data, illustrating its practical value.

Syftet med denna uppsats var att sammanställa forskning kring våldsutsatta kvinnors erfarenheter av vilka faktorer som påverkat deras beslut i att stanna eller lämna en våldsam relation, samt vad som har ökat deras förutsättningar att lämna relationen. Detta besvarades med forskningsfrågorna: “Vilka faktorer påverkar att en kvinna stannar kvar i eller lämnar en relation där det förekommer våld?” och “Vad ökar förutsättningarna för att lämna en relation där det förekommer våld?”. De vetenskapliga artiklar som användes till resultatet baserades på kvalitativa studier som intervjuat kvinnor som har erfarenhet av att utsättas för partnervåld och som antingen har valt att stanna i eller lämna relationen. En innehållsanalys användes för att analysera resultatet. De teorier som användes var intersektionalitet och normaliseringsprocessen, för att få en ökad förståelse för hur dessa teorier förklarar processen gällande förutsättningarna för ett uppbrott. Resultatet visade att beslutet att stanna i eller lämna relationen påverkas av bland annat kvinnornas tillgång till stöd, om de har barn och vilka känslor de har inför situationen. De faktorer som resultatet visade ökade förutsättningarna för att lämna är kvinnans ekonomiska förutsättningar att försörja sig själv och landets lagar och normer. Den slutsatsen som drogs var att det är högst individuellt vad som påverkar om en kvinna väljer att stanna i eller lämna relationen, men att ökade förutsättningar att lämna inte garanterar att kvinnan väljer att genomföra uppbrottet, då andra faktorer kan spela in. / The purpose of this essay was to compile research on abused women's experiences of which factors influenced their decision to stay or leave a violent relationship, as well as what they think has increased their chances of leaving the relationship. This was answered with the research questions: "Which factors influence whether a woman stays in or leaves a relationship where there is violence?" and "What increases the conditions for leaving a relationship where there is violence?". The scientific articles used for the results were based on qualitative studies that interviewed women who have experienced partner violence and who have either chosen to stay in or leave the relationship. A literature analysis was used to analyze the results. The theories that have been used are intersectionality and the normalization process, to gain an increased understanding of how these theories affect the process to leave the abusive relationship. The results showed that the decision to stay in or leave the relationship is influenced by, among other things, access to support, their children and what feelings they have about the situation. The results showed that conditions like the woman's financial conditions and the country's laws and norms increased the chance of leaving. The conclusion drawn is that it is highly individual what affects a woman to stay or leave the relationship, but that conditions that increase her opportunity to leave does not necessarily guarantee that the woman chooses to carry out the breakup, since other factors can play a part in her decision.

In this research the focus is on data-centric AI with a specific concentration on data quality evaluation and improvement for machine learning. We first present a practical framework for data quality evaluation and improvement, using a legal domain as a case study and build a corpus for legal argument mining. We first created an initial corpus with 4,937 instances that were manually labeled. We define five data quality evaluation dimensions: comprehensiveness, correctness, variety, class imbalance, and duplication, and conducted a quantitative evaluation on these dimensions for the legal dataset and two existing datasets in the medical domain for medical concept normalization. The first group of experiments showed that class imbalance and insufficient training data are the two major data quality issues that negatively impacted the quality of the system that was built on the legal corpus. The second group of experiments showed that the overlap between the test datasets and the training datasets, which we defined as "duplication," is the major data quality issue for the two medical corpora. We explore several widely used machine learning methods for data quality improvement. Compared to pseudo-labeling, co-training, and expectation-maximization (EM), generative adversarial network (GAN) is more effective for automated data augmentation, especially when a small portion of labeled data and a large amount of unlabeled data is available. The data validation process, the performance improvement strategy, and the machine learning framework for data evaluation and improvement discussed in this dissertation can be used by machine learning researchers and practitioners to build high-performance machine learning systems. All the materials including the data, code, and results will be released at: https://github.com/haihua0913/dissertation-dqei.

The main goal of this project, running at Neurofy AB, was that developing an AI recognition algorithm also known as, gas sensing algorithm or simply recognition algorithm, based on Artificial Intelligence (AI) technology, which would have the ability to detect or predict diary cattle diseases using odor signal data gathered, measured and provided by Gas Sensor Array (GSA) also known as, Electronic Nose or simply E-nose developed by the company. Two major challenges in this project were to first overcome the noises and errors in the odor signal data, as the E-nose is supposed to be used in an environment with difference conditions than laboratory, for instance, in a bail (A stall for milking cows) with varying humidity and temperatures, and second to find a proper feature extraction method appropriate for GSA. Normalization and Principal component analysis (PCA) are two classic methods which not only intended for re-scaling and reducing of features in a data-set at pre-processing phase of developing of odor identification algorithm, but also it thought that these methods reduce the affect of noises in odor signal data. Applying classic approaches, like PCA, for feature extraction and dimesionality reduction gave rise to loss of valuable data which made it difficult for classification of odors. A new method was developed to handle noises in the odors signal data and also deal with dimentionality reduction without loosing of valuable data, instead of the PCA method in feature extraction stage. This method, which is consisting of signal segmentation and Autoencoder with encoder-decoder, made it possible to overcome the noise issues in data-sets and it also is more appropriate feature extraction method due to better prediction accuracy performed by the AI gas recognition algorithm in comparison to PCA. For evaluating of Autoencoder monitoring of its learning rate of was performed. For classification and predicting of odors, several classifier, among alias, Logistic Regression (LR), Support vector machine (SVM), Linear Discriminant Analysis (LDA), Random forest Classifier (RFC) and MultiLayer perceptron (MLP), was investigated. The best prediction was obtained by classifiers MLP . To validate the prediction, obtained by the new AI recognition algorithm, several validation methods like Cross validation, Accuracy score, balanced accuracy score , precision score, Recall score, and Learning Curve, were performed. This new AI recognition algorithm has the ability to diagnose 3 different diary cattle diseases with an accuracy of 96% despite lack of samples. / Syftet med detta projekt var att utveckla en igenkänning algoritm baserad på maskinintelligens (Artificiell intelligens (AI) ), även känd som gasavkänning algoritm eller igenkänningsalgoritm, baserad på artificiell intelligens (AI) teknologi såsom maskininlärning ach djupinlärning, som skulle kunna upptäcka eller diagnosera vissa mjölkkor sjukdomar med hjälp av luktsignaldata som samlats in, mätts och tillhandahållits av Gas Sensor Array (GSA), även känd som elektronisk näsa eller helt enkelt E-näsa, utvecklad av företaget Neorofy AB. Två stora utmaningar i detta projekt bearbetades. Första utmaning var att övervinna eller minska effekten av brus i signaler samt fel (error) i dess data då E-näsan är tänkt att användas i en miljö där till skillnad från laboratorium förekommer brus, till example i ett stall avsett för mjölkkor, i form av varierande fukthalt och temperatur. Andra utmaning var att hitta rätt dimensionalitetsreduktion som är anpassad till GSA. Normalisering och Principal component analysis (PCA) är två klassiska metoder som används till att både konvertera olika stora datavärden i datamängd (data-set) till samma skala och dimensionalitetsminskning av datamängd (data-set), under förbehandling process av utvecling av luktidentifieringsalgoritms. Dessa metoder används även för minskning eller eliminering av brus i luktsignaldata (odor signal data). Tillämpning av klassiska dimensionalitetsminskning algoritmer, såsom PCA, orsakade förlust av värdefulla informationer som var viktiga för kllasifisering. Den nya metoden som har utvecklats för hantering av brus i luktsignaldata samt dimensionalitetsminskning, utan att förlora värdefull data, är signalsegmentering och Autoencoder. Detta tillvägagångssätt har gjort det möjligt att övervinna brusproblemen i datamängder samt det visade sig att denna metod är lämpligare metod för dimensionalitetsminskning jämfört med PCA. För utvärdering of Autoencoder övervakning of inlärningshastighet av Autoencoder tillämpades. För klassificering, flera klassificerare, bland annat, LogisticRegression (LR), Support vector machine (SVM) , Linear Discriminant Analysis (LDA), Random forest Classifier (RFC) och MultiLayer perceptron (MLP) undersöktes. Bästa resultate erhölls av klassificeraren MLP. Flera valideringsmetoder såsom, Cross-validering, Precision score, balanced accuracy score samt inlärningskurva tillämpades. Denna nya AI gas igenkänningsalgoritm har förmågan att diagnosera tre olika mjölkkor sjukdomar med en noggrannhet på högre än 96%.

Text in Afrikaans / In hierdie studie word gevra of vryheid nog 'n haalbare en nastrewenswaardige ideaal is aan die hand van Michel Foucault se werk Die vraag word ondersoek in die lig van die verskillende sienings van die subjek wat Foucault ontwikkel het, sowel as sy sienings oor mag en vryheid. Ter gevolgtrekking word bogenoemde vraag gekwalifiseerd bevestigend beantwoord, ook met verwysing na die Suid-Afrikaanse konteks, tydens en na Apartheid, en word kortliks gepoog om vryheid binne 'n Suid-Afrikaanse konteks te bedink. / In this study it is asked whether freedom is still an ideal that is attainable and worthy of pursuit with reference to the work of Michel Foucault. This question is investigated in the light of the different views of the subject which Foucault developed, as well as his views on power and freedom. In conclusion the abovementioned question is answered with qualified affirmation, also with regard to the South African context, during and after Apartheid, and a brief attempt is made to think freedom in a South African context. / Philosophy, Practical & Systematic Theology / M.A. (Wysbegeerte)

Approved for public release, distribution is unlimited / Wide local area networks (WLAN) are increasingly important in meeting the needs of next generation broadband wireless communications systems for both commercial and military applications. Under IEEE 802.11a 5GHz WLAN standard, OFDM was chosen as the modulation scheme for transmission because of its well-known ability to avoid multi-path effects while achieving high data rates. The objective of this thesis is to investigate the performance of the IEEE 802.11a WLAN standard receiver over flat fading Nakagami channels in a worst case, pulse-noise jamming environment, for the different combinations of modulation type (binary and non-binary modulation) and code rate specified by the WLAN standard. Receiver performance with Viterbi soft decision decoding (SDD) will be analyzed for additive white Gaussian noise (AWGN) alone and for AWGN plus pulse-noise jamming. Moreover, the performance of the IEEE 802.11a WLAN standard receiver will be examined both in the scenario where perfect side information is considered to be available (optimum receiver) and when it is not (sub-optimum receiver). In the sub-optimum receiver scenario, the receiver performance is examined both when noise-normalization is utilized and when it is not. The receiver performance is severely affected by the pulse-noise jamming environment, especially in the suboptimum receiver scenario. However, the sub-optimum receiver performance is significantly improved when noise-normalization is implemented. / Lieutenant, Hellenic Navy

Resiliency has become common in child welfare parlance in recent decades and producing resilient youth is touted as the panacea to improving notoriously poor outcomes for youth in care, when compared to youth not in the care of the state. The Looking After Children (LAC) system emerged in the U.K out of neoliberal and managerial policies of the 1990s. The LAC system, and its corresponding Assessment and Action Record (AAR), was subsequently imported to Canada and has been heralded to foster resilience in youth in care. The AAR is composed of hundreds of tick box questions posed to young people in care, child welfare workers, and foster parents; these questions are pedagogical and the mined data from the AAR is aggregated to inform child welfare policy. The Looking After Children: A Practitioner’s Guide (Lemay & Ghazal, 2007) instructs workers how to administer the AAR, Second Canadian adaptation (AAR- C2), and it informs workers how to do their job. The notion of resilience in the Practitioner's Guide and the AAR-C2 are based in normative development and day to day experiences (Lemay & Ghazal, 2007). My interest in the LAC system emerges out of my experiences as a child welfare worker and my experience of being a youth in care. I wondered how it was, given the oppressive track record of child welfare in Canada, that the state could initiate a system to produce normal youth. This was a particularly salient question given the massive over- representation of Indigenous youth in foster care. With this critical curiosity as a point of departure I employed a Foucauldian inspired discourse analysis of the Looking After Children: A Practitioner’s Guide (2007, University of Ottawa Press), and three versions of its corresponding Assessment and Action Record, Second Canadian adaptation (AAR- C2) (2006, 2010, 2016, University of Ottawa). My analysis asked the question: How have we come to this ideal of resiliency? What were the contingencies and complex set of practices that enabled this specific notion of resilience to emerge in child welfare? What are the material outcomes of this notion of resilience? My findings suggest that: Youth in care are produced as deviant and outside of normal development, versus the desired resilient youth; youth in care and foster parents are responsibilized to produce resilient outcomes, which can never actually be achieved; the AAR-C2 acts as a surveillance system to enable to production of neoliberal subjects; the LAC system and the AAR-C2 are a method of colonization of Indigenous youth in care. / Graduate

S’inscrivant dans le cadre d’une étude pancanadienne portant sur la normalisation du cannabis et la stigmatisation du tabac, le principal objectif de cette recherche était de comprendre quelle place occupe l’évaluation des risques dans la compréhension des perceptions et comportements liés aux deux substances. Pour ce faire, à partir d’un devis mixte – quantitatif et qualitatif, nous avons 1) décrit les perceptions et opinions des participants quant à leur propre consommation de cannabis et/ou de tabac et quant à la consommation que font les gens en général des mêmes substances. Nous avons aussi 2) décrit comment les participants évaluent les risques liés à chacune des substances. Enfin, sachant que la perception du risque est intimement liée au comportement d’un individu, nous avons cherché 3) à préciser comment l’évaluation du risque agit sur les comportements et les opinions des quelques 50 participants, hommes et femmes âgés entre 20 et 49 ans, bien intégrés socialement, envers le tabac et le cannabis. Il s’avère que les fumeurs de cannabis, qui ont insisté sur la distinction à faire entre la manière dont ils font usage de la substance et une consommation abusive, valorisait le contrôle que leur permettait, selon eux, la consommation de cannabis. La consommation de cigarettes, quant à elle, était perçue négativement pour des raisons opposées puisque de l’avis des participants à l’étude, elle engendrerait chez le fumeur une incapacité à se maîtriser et un besoin compulsif de fumer. Dans cette optique, les risques liés au cannabis étaient perçus, par la plupart, comme étant contrôlables, à l’exception du jugement d’autrui qui demeurerait incertain et sur lequel il serait impossible d’avoir du contrôle. La réaction de certaines personnes de leur entourage étant ou bien imprévisible ou négative, c’est ce qui les amènerait à fumer principalement en privé. Le contrôle social formel aurait finalement peu d’influence étant donné le fort sentiment qu’ont les répondants qu’il ne s’appliquera tout simplement pas à eux. / In the context of a pan-Canadian study pertaining to the normalization of cannabis and the stigmatization regarding tobacco, the main objective of this research was to understand the importance given in the evaluation of the risks in the understanding of the perceptions and behaviors as they relate to the aforementioned substances. Based on a mixed method, qualitative and quantitative, the process undertaken was 1) to describe the perceptions and opinions of respondents regarding their own use of cannabis and/or tobacco as well as the use of the same substances by the general public. We then 2) described how the respondents evaluate the risks as they relate to each substance. Finally, knowing that the perception of the risks are intimately related to the behavior of an individual, we 3) sought to determine how the evaluation of risks affects the behaviors and opinions of the 50 study participants, all of whom are socially well adjusted men and women between the ages of 20 to 49, regarding tobacco and cannabis. The cannabis smokers insisted on the importance of making a distinction between the way they use the substance versus an abusive use. They ascertained that there was a degree of control regarding the use of cannabis. The use of tobacco was perceived negatively for opposite reasons as it created a dependency to the smoker, an incapacity to control its use and the creation of a compulsive urge to smoke. Given these optics, the risks related to use of cannabis were perceived by most as controllable with the exception of judgment from others which remained uncertain and which would be impossible to control. The reaction of certain people within their surroundings was either unpredictable or negative which lead them to smoke mainly in private. Finally, formal social control would very little influence on the respondents given the strong feeling they had that this did not apply to them.