Carleton University Institutional Repository
Not a member yet
20816 research outputs found
Sort by
Spatial Analysis of Core Housing Needs in Ottawa Between 2016 and 2021
No full textIn this research, I explore core housing needs indicators in Ottawa between 2016 and 2021. This study comprises two analyses: spatial pattern analysis and spatial relationship analysis. The analyses reveal a clustered pattern, although it resembles a random pattern with only a slight increase in the clustering pattern. The downtown core and its surroundings exhibit the highest percentage of these indicators. Significant differences in the characteristics of the downtown core compared to the outer suburbs are evident in their response to unaffordability issues. These findings underscore two points: Firstly, they highlight the importance of analyzing housing inequality at a local scale using a spatial perspective. Secondly, the findings illustrate the dynamic nature of unaffordability and emphasize the necessity to redefine this to address nuanced differences that may occur within cities. This research contributes to understanding core housing needs by proposing three distinct types: voluntary unaffordability, conscious unaffordability, and compelled unaffordability
ABCD and Template Fitting for Background Estimation Using 90 Signal Models in the Search for Emerging Jets at the ATLAS Experiment at the Large Hadron Collider
No full textAnalysis is presented as part of the search for the physical signature known as an emerging jet in the ATLAS detector at the Large Hadron Collider. This signature occurs in a proposed dark sector, which interacts with the Standard Model via a dark mediator particle, Xd. This interaction may occur at displaced points from the original interaction point, resulting in the sudden emergence of these jets. The analysis uses Monte-Carlo simulated events at a luminosity of 139fb−1 and an energy of √s = 13TeV . This thesis shows the background estimation and sensitivity to the theoretical cross-section of 90 Emerging Jets signal models found by a data-driven ABCD method. This is done for a model independent approach and a model dependent approach that uses machine learning to create the input ABCD planes. Also presented is a similar study using an MC-driven template fitting method for background estimation and sensitivity calculations
Gathering, Governing, and Gifting Food: Community Economy and Food Distribution in the First Nation of Na-Cho Nyäk Dun
Get PDFThis research with the First Nation of Na-Cho Nyäk Dun (NND) explores how customary food practices and potlatch traditions might inform community-oriented food distribution and food governance. Food and potlatch practices contribute to NND’s community economy – the everyday relationships, activities, and decisions that sustain people and land. Grounded in feminist and decolonial community-engaged methodologies, this thesis integrates diverse community economies, Indigenous food sovereignty, and gift economy concepts. Community interviews emphasize the multidimensional values of food. Food and potlatch traditions generate insights into healing multiple relations by focusing on community strengths and the power of food as a convener of people and place, of human and more-than-human, of knowledge and skills, and of past, present, and future generations. Strategic, community-informed recommendations are grouped into themes of gathering (with the land and together), governing (for well-being, rematriation, and a circular food economy), and gifting (to honour food and knowledge as sacred)
Water and Multisensory Experiences of Architecture: Eighteenth–Century Ottoman Fountains in Travel Literature
No full textThis study investigates the multisensory experiences of eighteenth–century Ottoman public square fountains. Through Ottoman poetry, I establish an existing tradition of writing on architecture through water. As Europeans travelled to Istanbul, they encountered this cultural practice. Thus, I present the first two chapters as historico– cultural foundations of sensory explorations into eighteenth–century fountains, building to critical analyses in the final chapters. In the third chapter, I present Julia Pardoe’s 1838 The Beauties of the Bosphorus as a vital female sensory experience in nineteenth–century travel literature. In the final chapter, I critically engage with William Henry Bartlett’s drawings for Pardoe’s book, and Antoine–Ignace Melling’s 1807–1824 Voyage pittoresque de Constantinople et des rives du Bosphore to explore prints as objects of multisensory experiences. Thus, I reveal the interconnected emotional and physical experiences of everyday people in architectural settings, offering new avenues for exploring connections amongst those of the late Ottoman period
Metric-Based Frame Selection and Deep Learning Model With Multi-Head Self Attention for Classification of Ultrasound Lung Video Images
Get PDFDetection of COVID-19 manifestations in lung ultrasound (US) images has gained attention in recent times. The current state-of-the-art technique for distinguishing a healthy lung from COVID-19 infected or bacterial pneumonia infected lung uses non-adjacent frames or equally spaced frames from the video. However, the frame content or correlation between the selected frames has not been taken into consideration for frame selection. In this paper, a metric-based frame selection approach is proposed for three-way classification of lung US videos, and the influence of the frame selection method on image classification accuracy is studied. A deep learning model comprising of a pre-trained model (VGG16) for feature extraction, multi-head attention for feature calibration, global averaging for feature reduction, and a dense layer for classification is proposed. The pre-trained model is re-trained using cross-entropy loss with balanced weights to handle class imbalance. Two types of classification approaches are considered: i) few frames in a video are selected using the proposed metrics; and (ii) all frames in a video are considered. With VGG16 as the pre-trained model, a mean balanced sensitivity of COVID-19, bacterial pneumonia, and healthy classes with 0.82, 0.89, and 0.87, respectively was achieved using 5-fold cross-validation. The results show that even random selection of frames performs better than fixed frame selection and the proposed frame selection method outperforms the state-of-art fixed frame selection irrespective of the type of backbone model used for lung US classification
Towards Characterisation and Classification of Canadian Macrotidal Salt Marshes
No full textDespite interest in understanding the extent, distribution, and condition of tidal marshes in Canada, they have yet to be comprehensively mapped. Existing inventories show significant overestimation of Canadian tidal marsh extent, indicating that a regional model may be required. This thesis provides a review of freely available remote sensing data relevant to tidal marshes in the Bay of Fundy and uses freely available medium-resolution imagery to classify high and low tidal marsh extent in the Cumberland Basin for 2020 and 2023. Prediction maps are compared to five existing global and regional tidal marsh datasets and used to generate predicted change (activity) data. Assessment of prediction maps and activity data are then used as the basis to discuss the optimal sensor(s) and minimum requirements, the effects of model optimisation, and the potential for using medium-resolution imagery for the generation of activity data operationally for carbon inventories
Tequio y Tierra: Regenerative Architecture in Oaxaca
No full textThis thesis explores regenerative materials and earth construction in marginalized Oaxaca, Mexico. The Mezcal industry – which has exploded in recent years due to celebrity-fueled popularity – is causing serious ecological damage due to industrial-scale monoculture and improper disposal of its plant-base by-products bagasso and viñaza. Combined, these alternative materials can create circular construction and economies, giving agency to vulnerable sectors of the population while responding to the global ecological crisis. To destigmatize earth construction, this thesis proposes a centre for research, education, and innovation, for Earth and Mezcal’s by-products. The proposition is a network strategy for the handling, processing, and redistribution of Mezcal by-products into the local communities as new building materials. This thesis addresses pressing and current environmental, economic, and societal concerns through the lens of a specific place and its controversial by-product, striving to reveal how cities can become sustainable material mines for the future
Fight and flight: Behavioural drivers and evolutionary consequences of resource competition in birds
No full textAnimal competition determines the allocation of resources required for individual survival and reproductive success. These consequences of competition drive selection for traits that confer competitive advantages. While previous studies have quantified effects of fundamental behaviours on competition, research on more complex competitive behaviours has only recently emerged due to improvements in data collection and analysis methods. My research aims to advance our understanding of how resource competition drives the evolution of extreme flight anatomy and how complex flight and social behaviours influence competition in birds, a group of species that vary in competitive abilities. First, I investigate the evolution of flight apparatus morphologies related to high-performance flight in hummingbirds through the lens of species foraging competition. I show that variation in mass-adjusted keel and humerus morphologies is not explained by species differences in foraging strategies, but instead explained by sex differences presumably driven by within-species competition. Next, I assess the intrinsic drivers of unpredictable flight in hummingbirds, a complex behaviour expected to facilitate success during resource competition. I find that species with the most unpredictable flight demonstrate strong transitional and rotational maneuverability performance, while maneuverability itself has little to no effect on moment-to-moment changes in unpredictability. Instead, I show that unpredictable flight is achieved during slow flight, highlighting the nuanced relationships among intrinsic drivers of complex locomotor behaviours. Then, I examine the influence of social behaviour on the outcome of among- and within-species resource competition in North American birds. I show that more social species are weaker competitors than less social species, but demonstrate increased competitive success when in the presence of other conspecifics, all at the cost of increased conspecific competition. Finally, I discuss alternative ecological and evolutionary factors that may drive differences in resource competition abilities among birds, and highlight opportunities to integrate modern experimental approaches in future research to enrich our understanding of the dynamics of avian resource competition. Overall, my thesis research expands our knowledge of the evolutionary consequences of competition on extreme morphologies, the biomechanical facilitators of competitive locomotor behaviour, and the influence of social behavioural strategies on the outcome of resource competition
'What is in a Name?’: Using Natural Language Processing Techniques to Examine Attitudes About Solitude
No full textThis research characterized how adolescents and young adults describe someone who ‘enjoys solitude’ – and how these descriptions relate to their experiences of solitude. Participants’ descriptions of someone who enjoys solitude were coded using content analysis and lexicon-based sentiment analysis (from natural language processing). The content analysis found five themes: Introvert, Ambivert, Neutral, Positive, and Negative, representing different attitudes participants held towards one who enjoys solitude. The themes were differentially associated with attitudes towards solitude, and time alone alongside valence, arousal, and dominance scores from the sentiment analysis. Expectedly, compared to adolescents, emerging adults displayed more positive views of someone who enjoys solitude, evidenced by both methods. This research improves understanding of the utilities of language as indices of internal processes and provides insight into a novel method for psychological research. Most notably, results empirically support developmental differences in the perception of solitude between adolescents and emerging adults
Indoor Radio Dot Placement Optimization using UE Positioning and K-Means Clustering
No full textThis research evaluates the performance impact from dynamic information, mainly user density and distribution, to quantitatively evaluate radio dot adjustment algorithms that can better accommodate for cost-effective performance solutions. The number of UEs and their distribution are simulated, with the Machine Learning (ML) cluster algorithm of K-means being used to evaluate the ideal scenario where all the RD unit locations are adjusted. Further thesis specific algorithms are used to improve network performance for a cost-efficient solution is implemented. Results have proved that dynamic information is one of the key factors with major impact to the network performance. Adjusting RD unit placements by taking the dynamic information into account could provide a cost-efficient solution to optimize the Indoor network performance