39568 research outputs found
Sort by
Confronting Acoustic Data Scarcity: A Physics-Informed Approach to Trustworthy Machine Learning in Canada's Arctic
No full textThis thesis presents a novel, data-centric framework for building trustworthy artificial intelligence for passive acoustic monitoring in the Arctic, a region increasingly challenged by climate change and geopolitical pressures. The research addresses model brittleness in ship-radiated noise classification, framing it as a data problem that can be solved through a systematic, iterative process of data exploration, diagnosis, and augmentation.
A detailed analysis of ship-radiated noise using horizontal line array element data provides a characterization of the complex variability of acoustic signatures. The thesis quantifies the horizontal directionality of radiated noise from individual ships and the broad-scale impact on the ambient soundscape, leveraging a unique data opportunity presented by the COVID-19 pandemic. Using these insights, a human-in-the-loop methodology is developed to diagnose the specific failure modes of a custom deep neural network classifier. This is achieved by visualising how real-world variability, such as source-receiver range and operational state changes, manifests in the model's learned feature space.
The thesis culminates by demonstrating a physics-informed data augmentation strategy as the solution to data scarcity and diagnosed failures. Through the generation of targeted, high-fidelity synthetic data, this approach measurably improves classifier robustness on unseen real-world data, providing a validated methodology for developing reliable automated passive acoustic monitoring systems in complex environments
Investigation of Gas Evolution and Safety of Materials for Lithium and Sodium-Ion Batteries
No full textThe increased use of rechargeable batteries is one strategy in the fight against climate
change, as a higher battery usage allows for the integration of more renewable energy
sources. However, the development of batteries, particularly sodium-ion batteries, is still
underway, with improvements to energy density and lifetime as the main research goals.
Changes to the cell chemistry, the use of electrolyte additives, alloying negative electrode
materials, and increasing the upper cut-off voltage to achieve higher cell capacity are all
viable options to improve cell performance.
This work explores electrolyte additives and Pb as a negative electrode material in sodium
ion cells through on-line electrochemical mass spectrometry to investigate how these
components impact the gases that are produced during battery operation. The electrolyte
additives sodium difluorophosphate and 1,3,2-dioxathiolane 2,2-dioxide were studied in
comparison to a control electrolyte through half cell tests and storage tests with notable
differences in the gases evolved. Pb was compared to hard carbon and a blended electrode
containing both materials to study the differences in carbonate and ether based electrolytes.
Lastly, accelerating rate calorimetry was used to characterize the safety limits of a layered
oxide positive electrode material for lithium-ion cells. Various upper cut-off voltages were
used to probe the safety limitations of the material. Further investigation into the thermal
response of the layered oxide positive electrode was completed using x-ray diffraction
analysis. Overall, this work provides results for the gases produced from various cell
chemistries in sodium-ion batteries, as well as the thermal responses seen in lithium-ion
positive electrodes
COMPARING PERCEIVED AND UNMET MENTAL HEALTH NEEDS AMONG LGB AND HETEROSEXUAL INDIVIDUALS IN CANADA
No full textObjectives: This study aimed to (1) estimate and compare proportions of perceived and unmet mental health needs among Lesbian Gay, Bisexual (LGB) and heterosexual individuals in Canada; (2) assess changes in these needs between 2012 and 2022; and (3) evaluate associations between LGB identity and mental health needs.
Methods: Data derived from Canadian Community Health Survey - Mental Health (CCHS-MH) 2012 and Mental Health and Access to Care Survey (MHACS) 2022. Survey-weighted proportions were calculated, and multivariable logistic regression examined associations between LGB status and perceived or unmet needs, adjusting for covariates. Effect modification by age, sex, and race was tested.
Results: LGB adults reported higher perceived needs in 2012 (44% vs. 17%) and 2022 (62% vs. 23%) (p < 0.05). Unmet needs remained constant at 50%. Adjusted ORs for perceived need were 2.17 (2012) and 2.88 (2022). Unmet needs were significantly higher in 2012 (OR: 1.53) but not in 2022 (OR: 1.09). No effect modification was observed.
Conclusions: Despite progress among heterosexual individuals, unmet needs among LGB adults remain unchanged, highlighting the urgency for tailored, trauma-informed, sexual minority-affirming mental health interventions
UNDERSTANDING TRAVEL ACTIVITY IN THE DIGITAL ERA: MODELING TRAVELLER PROFILES AND INTERACTION BETWEEN ICT ACCESS, AND WORK ARRANGEMENTS
No full textAdvances in digital technology have changed how people work, socialize, and travel, yet most transportation planning still focuses on physical trips. This thesis explores how digital access and flexible work arrangements shape mobility and how transportation systems can adapt. It uses the 2024 Calgary CanTRAC Survey, which collected travel diaries and information on ICT ownership, online activity, and lifestyle preferences from 1,474 individuals. A two stage calibration corrected selection bias. Using this weighted dataset, k prototypes clustering identified three groups: Traditionalists, Hybrid Workers, and Active Professionals. A latent segmentation mixed logit model showed that technology access supports hybrid work, while job type and flexibility influence technology adoption. It introduces new ways to integrate digital participation into travel analysis and provides policy guidance to support flexible transit, digital inclusion, and more inclusive urban mobility
Effects of Shipping Traffic On Marine Microbial Community Composition
No full textShipping traffic across the world’s oceans continues to increase, and with it discharges of hydrocarbons and nutrient rich wastewater. In this thesis, we used a dataset of 971 marine metagenomic samples and estimates of shipping traffic density from satellite altimetry detections to measure the impact on the marine microbiome. Whole genome shotgun sequences were given taxonomic assignments through k-mer matching and aligned to genes for putative hydrocarbon degrading enzymes identified from ocean gene catalogs. Ship density had a significant (p < 0.05) impact on taxonomic community composition as measured by beta diversity metrics, but the effect size was smaller than other environmental variables such as temperature, indicating it did not impact the most abundant taxa. The abundance of several species of cyanobacteria had a significant association with ship density in more than one cruise, but the directions of association were not consistent with the expected effects on nitrogen fixing and non-nitrogen fixing genera from wastewater eutrophication. Five species belonging to known hydrocarbon degrading genera increased in abundance with ship density in individual sampling cruises. Additionally, gene sequences associated with several hydrocarbon degrading gene families increased in abundance with ship density in several individual cruises, and the aromatic degrading monoaromatic dioxygenase was found by a mixed linear model to increase with ship density across cruises. These taxonomic and functional results supported the hypothesis that operational discharges of oil from ships would increase the abundance of hydrocarbon degrading taxa. Overall, this thesis found that ship traffic density had a significant impact on microbial community composition, among this being associated with the increased abundance of hydrocarbon degrading taxa
VTSECURE: EARLY WARNING RANSOMWARE DETECTION AND REPORTING SYSTEM VIA VIRUSTOTAL FOR NON-EXPERT USERS
No full textMasters Thesis (Computer Science) on ransomware detection via web-based detection tools.Ransomware is a continuously growing cybersecurity risk to governments and organizations in critical sectors such as power grids, health care, and the banking/finance industry where data
privacy and system availability are essential for daily operations. Unlike other commonly known cybersecurity threats, ransomware attacks will not only disrupt operations but also
threaten data confidentiality and integrity.
To combat ransomware and its harmful effects, early detection systems can be developed and used by organizations and everyday users to identify
potential threats before they escalate, hence minimizing the likelihood of successful attacks.
This thesis proposes an early warning detection tool that leverages and examines files and URLs for potential threats. This research aims to create a tool for non-expert users, alerting them to harmful websites using VirusTotal’s API. By notifying users before they access malicious websites or download infected files, this tool can help prevent attacks from being triggered in the first
place
Healthy Sleep for Healthy Schools: Evaluating the Usability, Effectiveness, and Implementation Potential of a School-based Sleep Education Program
No full textAdolescence is a critical time of development; however, many adolescents are not obtaining sufficient sleep, an influential factor in well-being. There are few accessible
evidence-based sleep-education programs, leading to an “evidence-to-practice gap.”Schools are an optimal environment to teach sleep education, as this allows for many adolescents to learn about healthy sleep, in a place that is easily accessible to them. This dissertation evaluates the usability, effectiveness, and implementation potential of a school-based sleep education program, Healthy Sleep for Healthy Schools. A usability study (Study 1; n = 8) gathered quantitative and qualitative feedback from teachers based on Morville and Sullenger’s User Experience Honeycomb framework (2010). Quantitative feedback was neutral to positive, suggesting teachers found the program
useful, useable, findable, desirable, accessible, credible, and valuable. Qualitative feedback was mostly positive, with constructive feedback regarding program length, scheduling challenges, and interactivity of student content. HS4HS was evaluated for effectiveness with students (Study 2; n = 203), using a pre/post design. The study assessed insomnia symptoms, sleep habits, daytime sleepiness, knowledge about sleep, attitudes and beliefs about sleep, and psychosocial functioning. Findings demonstrated improvements in insomnia symptoms for students with subthreshold to severe insomnia, however no improvements were made for the overall sample. Qualitative feedback on program satisfaction suggested that students learned about sleep and how to improve sleep habits, with students suggesting modifications to improve engagement, delivery time, that the program should not replace physical education, and the study measures were burdensome. The third study examined the implementation potential of the HS4HS program (Study 3; n = 12). Semi-structured interviews were conducted with teachers and school stakeholders based on the RE-AIM implementation framework (Glasgow et al.,1999), to evaluate the reach, effectiveness, adoption, implementation, and maintenance of the HS4HS program. Feedback suggested aligning with the curriculum, and fostering buy-in and engagement. This dissertation research was undertaken to better understand the program and to make modifications based on end-user data. Further effectiveness testing will be conducted upon program modification, with the goal of eventually helping to close the “evidence-to-practice” gap and thereby improve adolescent health
EXPLORING EXPERIENCES OF A PILOT INTERGENERATIONAL PLAY PROGRAM IN NOVA SCOTIA, CANADA.
No full textPhysical inactivity and social isolation are two pressing public health concerns in Nova Scotia, particularly among older adults. Intergenerational play (IGP) has emerged as a promising community-based strategy to address these issues by promoting both physical activity (PA) and social connection (SC). This thesis describes a study designed to explore the factors that influence participation in “Free for All” (FFA), a locally developed IGP program modeled after ParkPlay in the United Kingdom. FFA offered one free drop-in session in Halifax during the summer of 2024. Guided by a constructivist worldview and framed by the Social Ecological Model (SEM), this study employed a qualitative design to understand participant experiences. Data were collected through semi-structured interviews with participants to explore personal experiences, barriers, and facilitators to participation in FFA. Data were analyzed using reflexive thematic analysis, and three themes were generated: 1) Negotiating entry into play: ‘on ramps’ matter, (2) Play as an intergenerational equalizer: children open doors and adults walk through them, and (3) The fragile infrastructure of belonging: material conditions build social connections. The findings of this study contribute to a growing body of knowledge that IGP can serve as an inclusive, community-based health promotion strategy to promote PA and SC. Findings highlight the importance of cost-free design, child-led openness and supportive facilitation to promote participation. They also suggest that material and environmental conditions, such as accessibility and program continuity, are critical for sustaining participation and connection. Together, these insights have created a framework to implement IGP in Nova Scotia