39568 research outputs found
Sort by
LATENT STRUCTURE IDENTIFICATION AND PERSONALIZED VARIABLE SELECTION
No full textThe identification of latent structures and the selection of personalized variables are critical to enhance the interpretability of models, predictive performance, and decision-making efficiency in complex data environments. There are three parts in this thesis.
In the first part, we focus on the development of a method for identifying the latent spatial patterns for a response variable in spatial data. We propose a new method that calculates the similarity scores between different locations with a supervised random forest model to effectively capture spatial dependencies of a response variable. The similarity score is derived from the proportion of trees in which two locations fall in the same terminal node for the same values of other predictors. This resulting similarity matrix is then used to derive eigen-scores and spatial clusters, which significantly improve the performance of models such as XGBoost, GWR, and random forest in both simulations and real datasets.
In the second part, we develop an effective neural network pruning method based on backwards LASSO selection that can simultaneously select
features and structure. We show that the LASSO shrinkage problem in neural networks can be re-written as a standard weighted regression or classification problem with LASSO penalty. Our proposed method starts from a dense neural network which contains all structures without feedback, and prunes links to select the optimal sparse neural network structure. The results of this structure selection highlight the inadequacy of commonly-used feedforward structures. By examining the selected structure, we are able to gain insight into the linear or nonlinear properties of the estimated function, and thus better interpret the underlying function.
Finally, personalized variable selection is a novel topic to address an important problem. In many real-world applications, some variables may be costly or difficult to obtain. For example, in healthcare, ordering excessive medical tests can lead to unnecessary expenses, long waiting times, and patient discomfort. In the personalized variable selection paradigm, we consider the problem of using a fitted model to make predictions for a new observation where we have not yet measured all these costly variables. We assess the predictive value of the potentially useful predictor variables for this new observation, in order to decide which predictors are worth measuring for this observation. We introduce a novel metric called the Expected Loss Improvement Estimate (ELIE), which quantifies the expected gain in predictive accuracy from measuring a missing variable. The core idea of our method is that large ELIE suggests greater variability in predictions, indicating that collecting the true values of the missing variables is highly valuable for those data points. This approach can help us determine when imputation is sufficient and when additional data collection is necessary to maximize model performance
Introducing the Substituted Azobispyrrole Framework: Synthesis and Properties
No full textAn unprecedented azobispyrrole framework is introduced, wherein two aryl-substituted pyrroles are joined by an azo (-NN-) linkage. Functionalisation was demonstrated via pyrrolic N-methylation and by -BF2 complexation using the coordinating abilities of the pyrrolic and azo nitrogen atoms. Control of co-planarity enables tunability with maximal absorption and emission spanning almost 300 nm
A Sense of Home: Mi’kmaq and Relocation Policies in the Early Twentieth Century
No full textThe Department of Indian Affairs formally enforced its centralization policy from 1942 to 1949 with the goal of removing all Mi’kmaq living in over twenty communities in Nova Scotia to two reserves. This thesis explores Mi’kmaw concepts of home and homeland between the seventeenth and early twentieth centuries to better understand the Mi’kmaw experience of forced relocation. This thesis relies on language, stories and oral histories to unsilence Mi’kmaw voices in the historical narrative. It pays particular attention to the experiences and contributions of Mi’kmaw women to address a gap in the literature. The family was the heart of Mi’kmaw society and the wikuom, or wigwam, provided a home that allowed L’nu’k, The People, to connect to homelands. Mi’kmaw used petitions to assert political sovereignty and express ownership and property rights. Centralization failed but radically altered Mi’kmaw political structure, disrupting kin networks and connections to traditional homelands
Interplay Between Metabolic Alterations, Bipolar Disorders, and White Matter Microstructure
No full textAn investigation into the white matter microstructural alterations associated with obesity, obesity associated biomarkers, bipolar disorders, and psychiatric obesogenic medications.The goal of this dissertation was to describe where and to what extent obesity affects brain white matter in the general population and bipolar disorders. We used magnetic resonance imaging derived diffusion tensor imaging to investigate white matter microstructural alterations in the brain. We conducted voxel-based and region of interest meta-analyses, collaborated with the ENIGMA Bipolar Disorder Working Group to perform mega-analyses in large international samples of over 2000 participants, and explored deeper phenotyping through the Halifax Study. Our results show localized obesity associated white matter abnormalities in the general population. We found obesity, obesity related biomarkers, and psychiatric obesogenic medications added to the white matter alterations in bipolar disorder. Overall, this work has deepened our understanding of how obesity impacts the brain. We hope these insights draw attention to the hidden costs of obesity and provide new treatment options for prevention of negative brain, cognitive, and clinical outcomes
Development of a Point-of-Care Rapid Test for The Detection of Bovine Viral Diarrhea Antibodies in Bovine Blood
No full textBovine viral diarrhea negatively impacts animal health and producer profits in beef and dairy herds worldwide. There is no cure for BVDV, so strict disease management and preventative practices are needed to mitigate the adverse effects of the disease. Disease control practices require accessible and affordable BVDV diagnostic antigen AND antibody tests. While BVDV rapid lateral flow antigen tests are widely available, access to POC antibody detection tests remains limited, with poor diagnostic accuracy. Therefore, we developed a vertical flow rapid antibody test, utilizing novel technologies to detect BVDV antibodies in cattle blood. Thus, this project aimed to develop a novel, accurate, and affordable rapid POC test for the detection of BVDV antibodies in bovine blood. Conjugated HRP detectors and recombinant BVDV capture antigens were utilized to develop a vertical flow immunoassay. Optimization testing was performed to determine the best conditions for detecting the target BVDV antibodies, yielding a test that could detect antibody titers as low as 1:4 in bovine plasma in under 10 minutes.Bovine viral diarrhea virus (BVDV) is a highly contagious pestivirus responsible for many production-limiting symptoms in cattle worldwide. There is no cure for BVDV, and management is highly dependent upon preventative care, immunization, and establishment of herd immunity. A point-of-care (POC) BVDV antibody detection test would allow producers to quickly and effectively determine their herd's immunity and infection status, greatly aiding BVDV management programs. Thus, this project aimed to develop a novel, accurate, and affordable rapid POC test for the detection of BVDV antibodies in bovine blood. Conjugated HRP detectors and recombinant BVDV capture antigens were utilized to develop a vertical flow immunoassay. Optimization testing was performed to determine the best conditions for detecting the target BVDV antibodies, yielding a test that could detect antibody titers as low as 1:4 in bovine plasma in under 10 minutes. A total of 195 positive and 52 negative plasma samples from both beef and dairy cattle were run on the vertical flow test (VFT) and a reference industry standard enzyme-linked immunosorbent assay (ELISA) was used to determine the test's diagnostic sensitivity (DSe) and specificity (DSp). The VFT had a DSe of 98.5% and DSp of 96.2%, while the reference ELISA had a DSe of 97.4% and a DSp of 96.2%. Comparatively both tests had substantial agreement (Cohen’s kappa = 0.91). The newly developed VFT has a diagnostic accuracy exceeding other reported rapid antibody diagnostic tests. Furthermore, the VFT’s DSe is comparable to in-lab ELISA results, suggesting that the test could be used to provide highly accurate pen-side results. The shortened performance time, ease of use, and ability of this test to be performed outside of a laboratory could greatly improve BVDV surveillance and management practices worldwide, reducing disease incidence and improving profit outcomes for beef and dairy producers alike
DIVERSIFYING BLACKNESS IN THE AGE OF EDIA
No full textEquity, Diversity, Inclusion and Accessibility (EDIA), has become a strategic priority for academic institutions in the Canadian Maritimes and beyond, as through them, universities hope to be seen as progressive. Race, gender and sexuality are just a few factors that can be addressed in EDIA initiatives, goals and objectives, usually implemented for a set time period. Intersectional identities such as race, class, sex, gender and citizenship status all play a role in informing the experience of being specifically a Black student in Canada, shaped by broader social understandings of race. This can result in the homogenization of differences within Blackness with a ‘one size fits all’ approach. Considering the nuances of identity politics that surround the experience of being a student, but also a racialized student within Canada, creates unique experiences and barriers that manifest when these and other identities overlap
Women on the Water: The Feminine Seascape Onboard Nova Scotia’s Nineteenth Century Deepwater Merchant Sailing Vessels
No full textSeafaring has long been considered as inherently masculine. This perspective has shaped the cultural legacy of Nova Scotian seafaring women, whose contributions to maritime industries are often forgotten or marginalized. This thesis challenges the dominant masculine seascape by establishing an equivalent feminine seascape. The feminine seascape recontextualizes ships as part of a woman’s household, a structure of power that bestowed married women with significant authority. By studying the life-writing of three Nova Scotian sea captains’ wives during the age of sail, as well as contemporary English-language newspapers and print culture, this thesis explores the paradoxical nature of women’s power within patriarchal systems, arguing that seafaring both challenged and reinforced ideals of Victorian married womanhood. Ideologies of domesticity provided sea captains’ wives with a discursive lens through which they could interpret maritime spaces and activities as normative for their phase of life. My thesis argues that this enabled captains’ wives to construct a multifaceted feminine seascape. It contributes to a growing field of gender history at sea and alters our understanding of captains’ wives. In doing so, it contributes to regional histories by positioning the experiences of rural, middle-class women as cosmopolitan and transnational, thereby disrupting the parochial and conservative narratives often attributed to, and occasionally embraced by, the Maritimes
Adapting Eco-Driving Feedback and Historical Visualization for Vessel Dashboards
No full textMaritime navigation is a significant source of greenhouse gas emissions. While large-scale cargo shipping is the major contributor, smaller maritime operations, including patrolling, fishing, public transit, and recreation, present unique challenges and opportunities for power management. Fuel consumption, power conversion, and environmental data can permit environmentally conscious and cost-effective decision-making when driving a boat. To achieve this, we need to understand how best to integrate such data into boat dashboard interfaces. In this work, we design an Eco Dashboard inspired by eco-driving feedback dashboards in the automotive industry, as well as a variant of the Eco Dashboard that additionally visualizes historical route and fuel consumption data (Eco + Historical Dashboard). In an experimental simulation (N = 30) involving 12 experienced mariners and 18 novices, we compared both interfaces with a typical boat dashboard that presented fuel and speed. Our findings suggest that dashboards incorporating historical data, alongside eco-driving features, improve fuel efficiency and decision-making, particularly for non-experienced users. The Eco Dashboard supported real-time adjustments during complex navigation, whereas the Eco + Historical Dashboard enhanced route planning and confidence in longer-term decisions. Participants also reported greater confidence and reduced cognitive load when using these systems. These results provide valuable insights for the future design of maritime dashboard systems, offering a pathway to more effective and environmentally conscious navigation tools
ANALYSIS OF SUSTAINABLE MINING SUPPLY CHAINS
No full textThe future of mining relies on solutions that are sustainable, efficient, and credible for the communities that live alongside mining operations. This research provides an integrated framework that addresses both ends of the spectrum: artisanal and small-scale mining (ASM), where immediate water conservation and basic treatment capabilities are crucial; large-scale mining operations, where decarbonized energy logistics within transparent governance structures are essential; as well as a unified decision analytics framework for conflict management in mining communities. This research addresses these gaps through four interrelated studies.. First, the review of sustainable operational efficiency (SOE) in the mining industry (MI). Second, the establishment of multi-stage water treatment networks that meet regulatory standards for reuse. Third, the development of transport modes for natural gas (NG) delivery to decrease the effects of traditional fuels. Finally, the implementation of commitment mechanisms that transform benefit-sharing and grievance processes from aspirations into realities. At the core of this framework are multi-objective optimization models that evaluates trade-offs aligned with the quadruple bottom line (QBL) of the sustainability framework. As well as decision analysis models for effectively translating stakeholder preferences into clear priorities through sensitivity assessments. Together, this dissertation contributes a cohesive field-ready methodology for designing mining systems for both ASM and large scale mining that preserve waterbodies, minimize emissions, and secure license to operate by aligning engineering design with responsible community-informed governance