Graduate Theses & Dissertations


Comparing Two Tablet-Based Visuomotor Tasks to Standard Laboratory Versions
The assessment of visuomotor function can provide important information about neurological status. Several visuomotor tasks exist for testing in the laboratory, although attempts to make these tests portable to allow quick and reliable assessment have been limited. We developed an assessment tool using two laboratory visuomotor tests as a tablet application: the double-step task, and an interception task. Performance was assessed by measuring the participants’ ability to reach toward unpredictably moving targets in each task. Response patterns were compared across equipment types to determine if participants were responding similarly to the moving targets in the standard laboratory and the tablet version of the tasks. On the double-step task, participants adjusted to the displaced target adequately in both the lab and tablet versions. On the interception task, participants intercepted non-accelerating targets, and performed worse on accelerating targets in both versions of the task. These findings suggest that the tablet version of these tasks assesses similar visuomotor processing as the respective laboratory version. Author Keywords: concussion assessment, double-step task, interception task, visuomotor processing, visuomotor system
Effect of Listing a Stock on the S&P 500 Index on the Stock’s Volatility
This paper investigates the effect of listing a stock on the S&P 500 Index on the stock’s volatility, using various econometrics models: GARCH and EGARCH. The study mainly addresses three issues; firstly, it analyzes stock volatility in two sub-periods, secondly, it determines whether the announcement can account for the fluctuations in the price of the stock, and finally, it investigates the change in the stock’s variance. After isolating the effects of external and industry shock by using the returns on the S&P 500 Index as a proxy, the author finds evidence of structural change in the volatility of stocks after that stock is added to the index. Additionally, the existence of a dominant symmetric effect, which captures the response of volatility to news, indicate that following the onset of including the stock on the index, information flowing into the market increased. However, the rate at which old news is captured in price falls. The empirical evidence also suggests that on average a stocks variance falls and that the announcement to list a stock on the index has little effect on the stock’s price. Author Keywords: EGARCH, GARCH, S&P 500 Index, Symmetric Effect, Volatility
Smote and Performance Measures for Machine Learning Applied to Real-Time Bidding
In the context of Real-Time Bidding (RTB) the machine learning problems of imbalanced classes and model selection are investigated. Synthetic Minority Oversampling Technique (SMOTE) is commonly used to combat imbalanced classes but a shortcoming is identified. Use of a distance threshold is identified as a solution and testing in a live RTB environment shows significant improvement. For model selection, the statistical measure Critical Success Index (CSI) is modified to add emphasis on recall. This new measure (CSI-R) is empirically compared with other measures such as accuracy, lift, efficiency, true skill score, Heidke's skill score and Gilbert's skill score. In all cases CSI-R is shown to provide better application to the RTB industry. Author Keywords: imbalanced classes, machine learning, online advertising, performance measures, real-time bidding, SMOTE
Mixed methods approaches in marine mammal science
This thesis explored the contribution of mixed methods approaches to marine mammal science through the use of concurrent and sequential designs to study distribution and feeding ecology of bowhead whales (Balaena mysticetus) in the Arctic region of Nunavik, Quebec, Canada. The study combines Inuit knowledge (IK), collected through semi-directed interviews with Inuit harvesters, and analyses of stable isotopes and trace elements (SI/TE) in baleen plates. A systematic literature review found that mixed methods are increasingly used in marine mammal ecology studies in remote locations, yet are still relatively rare and face a number of challenges. Both IK and SI/TE, indicated that bowhead whales have a seasonal pattern in their distribution and feeding ecology. They are most commonly present in productive nearshore areas in summertime, feeding in areas of great prey diversity, and moving to offshore areas in winter to fast. Mixed methods approaches used in this case study enabled the collection of complementary knowledge about bowhead whale ecology important for local management in a changing climate. This study also shows the value of mixed methods approaches for future marine mammal studies in Nunavik and elsewhere. Author Keywords: Arctic, bowhead whale, distribution, feeding ecology, mixed methods, traditional ecological knowledge
Modelling Submerged Coastal Environments
Built upon remote sensing and GIS littoral zone characterization methodologies of the past decade, a series of loosely coupled models aimed to test, compare and synthesize multi-beam SONAR (MBES), Airborne LiDAR Bathymetry (ALB), and satellite based optical data sets in the Gulf of St. Lawrence, Canada, eco-region. Bathymetry and relative intensity metrics for the MBES and ALB data sets were run through a quantitative and qualitative comparison, which included outputs from the Benthic Terrain Modeller (BTM) tool. Substrate classification based on relative intensities of respective data sets and textural indices generated using grey level co-occurrence matrices (GLCM) were investigated. A spatial modelling framework built in ArcGISTM for the derivation of bathymetric data sets from optical satellite imagery was also tested for proof of concept and validation. Where possible, efficiencies and semi-automation for repeatable testing was achieved using ArcGISTM ModelBuilder. The findings from this study could assist future decision makers in the field of coastal management and hydrographic studies. Keywords: Seafloor terrain characterization, Benthic Terrain Modeller (BTM), Multi-beam SONAR, Airborne LiDAR Bathymetry, Satellite Derived Bathymetry, ArcGISTM ModelBuilder, Textural analysis, Substrate classification Author Keywords:
SPAF-network with Saturating Pretraining Neurons
In this work, various aspects of neural networks, pre-trained with denoising autoencoders (DAE) are explored. To saturate neurons more quickly for feature learning in DAE, an activation function that offers higher gradients is introduced. Moreover, the introduction of sparsity functions applied to the hidden layer representations is studied. More importantly, a technique that swaps the activation functions of fully trained DAE to logistic functions is studied, networks trained using this technique are reffered to as SPAF-networks. For evaluation, the popular MNIST dataset as well as all \(3\) sub-datasets of the Chars74k dataset are used for classification purposes. The SPAF-network is also analyzed for the features it learns with a logistic, ReLU and a custom activation function. Lastly future roadmap is proposed for enhancements to the SPAF-network. Author Keywords: Artificial Neural Network, AutoEncoder, Machine Learning, Neural Networks, SPAF network, Unsupervised Learning
Novel Silica Sol-Gel Passive Sampler for Mercury Monitoring in Aqueous Systems
A novel passive sampler for mercury monitoring was prepared using organosilica sol-gel materials. It comprises a binding layer with thiol groups for mercury complexation and a porous diffusive layer through which mercury can diffuse and arrive at the binding layer. Our study demonstrated that this new sampler follows the principle of passive sampling. The mass of mercury accumulated in the binding layer depends linearly on the mercury concentration in solution, the sampling rate and the exposure time. A typical sol-gel sampler is characterized by a diffusive layer of 1.2 &mum, in which mercury ions diffuse with a coefficient of D = 0.09~10-6 cm2/s. The capacity for mercury uptake is approximately 0.64 &mug/cm2. Mercury diffusion and binding in the passive sampler are independent of the type of mercury-chloride complex. Its sampling rate increases with increasing water turbulence and decreases with increasing DOM amount. The field trial of sol-gel sampler in Miller Creek shows the concentration gained from the sol-gel passive sampler is slightly lower than that from the spot sampling. Author Keywords:
Scarring, sex assignment, and sex-specific sociality of Indo-Pacific humpback dolphins (Sousa chinensis) in the Pearl River Estuary and eastern Taiwan Strait
The Pearl River Estuary (PRE) and eastern Taiwan Strait (ETS) populations of Indo-Pacific humpback dolphins (Sousa chinensis) contain ~600 and 100 catalogued individuals, respectively. Population demography is important for conservation actions but few animals have been sexed by conventional methods. Longitudinal analysis of tooth-rake scars on dorsal fins showed scars disappear within 7 months; sexing adults by scarring is likely not impacted by juvenile scarring. Using dorsal fin photographs, sex assigned for 87% of catalogued PRE adults (n=300) and for 93% of ETS adults (n=60), using scars hypothesized from male-male competition, was in concordance with sex assigned by DNA, calf association, and ventral photographs. Scarring was higher in presumed males than females and in PRE females than ETS females. Female:male sex ratios were 3:2 (PRE) and 2:1 (ETS), though this likely results from biases in photo-identification methods. Social analysis with presumed sexes showed strong female-female associations in both populations but stronger female-male and male-male associations in PRE. These results support sex differentiation by scarring, which was a non-invasive approach, and sex assignment for many PRE and ETS individuals. Author Keywords: Indo-Pacific humpback dolphins, photo-identification, scarring, sex differentiation, sexes, social structure
Influence of nitrogen deposition on the vegetation community of Irish oak woodlands
In this study, the influence of N deposition on the vegetation community of semi-natural oak woodlands in Ireland was assessed through national and regional scale analysis of forest plot data. At both scales, Canonical Correspondence Analysis suggested that N deposition was a predictor of community composition, although site-specific soil characteristics were the strongest predictors of the species dataset. Threshold Indicator Taxon Analysis suggested that the vegetation community demonstrated the most change at 13.2 kg N ha-1 yr-1. While this change point falls within the current recommended critical load range for nutrient nitrogen for acidophilous oak dominated woodlands (10 to 15 kg N ha-1 yr-1), it is notable that 23% of species recorded had individual change points below this range, and could potentially be lost from this habitat if deposition increases. The results from this study suggest that, for acidophilous oak woodlands, habitat conservation policies should be unified with N emission reduction policies. Author Keywords: community composition, critical load, nitrogen depositioin, oak woodland, species richness, Taxon Indicator Threshold Analysis
Heavy Rydberg Photo-dissociation Cross-section Calculations and Experimental Progress Towards Cold Collisions in Lithium
This thesis is divided into two parts, each of which supports constructing and using a lithium magneto-optical trap for cold collision studies: Part I One outgoing channel of interest from cold collisions is the production of ion pairs. We describe an effective method for calculating bound-to-continuum cross-sections for charged binary systems by examining transitions to states above the binding energy that become bound when the system is placed within an infinite spherical well. This approach is verified for ionization of a hydrogen atom, and is then applied to the heavy Rydberg system Li+...I-. Part II A wavemeter previously built in the lab is redesigned for increased reliability and ease of use by replacing the optical hardware with a rocker system, which can be aligned in mere minutes rather than half a day as was previously the case. The new wavemeter has been tested through saturated absorption spectroscopy of lithium. Author Keywords: cross-section, dissociation, lithium, magneto-optical trap, Michelson, wavemeter
I investigated biodiversity of birds and vegetation associations along the boreal-tundra ecotone in Ivvavik National Park, Yukon Territory, and breeding adaptations used by American Robins (Turdus migratorius) at high latitudes. Twenty bird species were detected over three years using point-count surveys. Densities of American Robin, Dark-eyed Juncos (Junco hyemalis), and Yellow-rumped Warbler (Dendroica coronata) had positive relationships with tree and shrub density, whereas density of White-crowned Sparrows (Zonotrichia leucophrys) was negatively related to tree density. American Robins at this latitude raised only one brood, but females laid slightly larger clutches, the young fledged earlier, and pairs experienced higher nest-success than American Robins at more southerly latitudes. American Robins selected nest sites with high vegetation volume, at both the nest-site, and the nest-patch. This study is important for the first description of the bird community at this high latitude location, and describing how a species at the northern limit of the boreal forest has adapted to living with short-breeding seasons. Author Keywords: American Robin, Ivvavik National Park, Life History, Nest-stie selection, Northern limit
"Multimodal Contrast" from the Multivariate Analysis of Hyperspectral CARS Images
The typical contrast mechanism employed in multimodal CARS microscopy involves the use of other nonlinear imaging modalities such as two-photon excitation fluorescence (TPEF) microscopy and second harmonic generation (SHG) microscopy to produce a molecule-specific pseudocolor image. In this work, I explore the use of unsupervised multivariate statistical analysis tools such as Principal Component Analysis (PCA) and Vertex Component Analysis (VCA) to provide better contrast using the hyperspectral CARS data alone. Using simulated CARS images, I investigate the effects of the quadratic dependence of CARS signal on concentration on the pixel clustering and classification and I find that a normalization step is necessary to improve pixel color assignment. Using an atherosclerotic rabbit aorta test image, I show that the VCA algorithm provides pseudocolor contrast that is comparable to multimodal imaging, thus showing that much of the information gleaned from a multimodal approach can be sufficiently extracted from the CARS hyperspectral stack itself. Author Keywords: Coherent Anti-Stokes Raman Scattering Microscopy, Hyperspectral Imaging, Multimodal Imaging, Multivariate Analysis, Principal Component Analysis, Vertex Component Analysis


Search Our Digital Collections


Enabled Filters

  • (-) ≠ Burness
  • (-) = Master of Science
  • (-) ≠ Freeland

Filter Results


1974 - 2024
Specify date range: Show
Format: 2024/02/21