1722 research outputs found
Sort by
Poetika i imaginariji Montréala u suvremenoj kvebeckoj knjizevnosti
Get PDF[Poetics and Imaginaries of Montréal in Contemporary Québécois Literature.] Contemporary narratives no longer display a singular city but a palimpsest of cities. “L ’esprit migrateur” (Pierre Ouellet) and “la rencontre transculturelle” (Patrick Imbert) inhabit the new literary imaginary. What representations of Montréal do texts, such as La Québécoite by Regine Robin and La femme qui fuit by Anaïs Barbeau-Lavalette present? How can we describe the experience of wandering through the city of Montréal? And what are its relationships with transculturalism? If the topic of mobility has been common in Québécois literature since the 1980s, mainly in the works of migrant writers, mobility is nowadays not geocultural: it is rather symbolic and ontological.https://hrcak.srce.hr/22895
Access to Environmental Justice: A Manitoba Toolkit for Improving Public Participation; Final Project Report, February 2021
Get PDFThis final report details the implementation of the Manitoba Law Foundation funded project: Access to Environmental Justice: A Manitoba Toolkit for Improving Public Participation. The project materialized as a 6-part public legal education webinar series, delivered over the 2020/2021 academic year. The goal of the series is to increase Manitobans’ awareness of and participation in the decision-making processes that impact the environment in Manitoba and Canada.Funding for this project was provided by The Manitoba Law Foundation
A study on the modeling for obesity
Get PDFThis study primarily aims to develop an Agent Based Model (ABM) that can simulate the obesity rates based on statistical analysis and to find out how obesity is affected by risk factors in a Canadian environment. As obesity can have many causes, it is assumed that various risk factors, not just a decisive one, have an influence on obesity and they interact with one another. Therefore, unlike most previous studies, I approached the obesity problem as a Complex -Adaptive System (CAS). The data used for this study was provided by Statistics Canada, and the Canadian Community Health Survey (CCHS). This survey is a cross-sectional survey that collects self-reported information related to health status, health care utilization, and health determinants for the Canadian population. To build the Obesity ABM, it is necessary to find out which risk factors are closely associated with obesity and to what extent they interact with one another. Twelve categories of factors that are expected to influence the obesity rate were chosen on the basis of the related works. Through the statistical data analysis carried out, the main factors and variables for obesity were identified and their respective mathematical relationships obtained. From this, two categories that have several sub-factors for the obesity model were chosen. I implemented statistical data analysis on the CCHS dataset to see the interrelationship among the factors. Also, I implemented a year-to-year analysis that can show how people change their obesity status each year. Based on the data analysis result, I defined rules for how each risk factor changes each year. These rules are applied to the obesity model using NetLogo. The architecture of obesity model implementation consists of three main parts: The population module, the risk factor module, and the results module. Performance evaluation was conducted to examine whether the obesity model can simulate the obesity rate. For this evaluation, the data of CCHS from 2009 to 2014 and the result of the obesity model which is generated by simulation are compared. Model calibration was executed to fit the actual data to the model test result. The result of the model test shows that the percentage error is less than 5%. This means that the obesity model has high validity in predicting obesity for each risk factor. The obesity ABM is a useful tool to find out the risk factors related to obesity and their relationships in the Canadian population. Thus, this model can potentially assist to improve obesity management at various levels. At the individual level, everyone can find what kinds of strategies are best fit to improve her/his physical condition. Also, at a government or community level, it could help develop policies for people to continue to implement these strategies well. This will lead to reducing the associated social costs and help to promote national health.Master of Science in Applied Computer Scienc
Short Text Classification with Tolerance Near Sets
Get PDFText classification is a classical machine learning application in Natural Language Processing, which aims to assign labels to textual units such as documents, sentences, paragraphs, and queries. Applications of text classification include sentiment classification and news categorization. Sentiment classification identifies the polarity of text such as positive, negative or neutral based on textual features. In this thesis, we implemented a modified form of a tolerance-based algorithm (TSC) to classify sentiment polarities of tweets as well as news categories from text. The TSC algorithm is a supervised algorithm that was designed to perform short text classification with tolerance near sets (TNS). The proposed TSC algorithm uses pre-trained SBERT algorithm vectors for creating tolerance classes. The effectiveness of the TSC algorithm has been demonstrated by testing it on ten well-researched data sets. One of the datasets (Covid-Sentiment) was hand-crafted with tweets from Twitter of opinions related to COVID. Experiments demonstrate that TSC outperforms five classical ML algorithms with one dataset, and is comparable with all other datasets using a weighted F1-score measure.Master of Science in Applied Computer Scienc
Improving LULC Map Production via Semantic Segmentation and Unsupervised Domain Adaptation
Get PDFIn recent years, a lot of remote sensing problems benefited from the improvements made in deep learning. In particular, deep learning semantic segmentation algorithms have provided improved frameworks for the automated production of land-use and land-cover (LULC) map generation. Automation of LULC map production can significantly increase its production frequency, which provides a great benefit to areas such as natural resource management, wildlife habitat protection, urban expansion, damage delineation, etc. In this thesis, many different convolutional neural networks (CNN) were examined in combination with various state-of-the-art semantic segmentation methods and extensions to improve the accuracy of predicted LULC maps. Most of the experiments were carried out using Landsat 5/7 and Landsat 8 satellite images. Additionally, unsupervised domain adaption (UDA) architectures were explored to transfer knowledge extracted from a labelled Landsat 8 dataset to unlabelled Sentinel-2 satellite images. The performance of various CNN and extension combinations were carefully assessed, where VGGNet with an output stride of 4, and modified U-Net architecture provided the best results. Additionally, an expanded analysis of the generated LULC maps for various sensors was provided. The contributions of this thesis are accurate automated LULC maps predictions that achieved ~92.4% of accuracy using deep neural networks; production of the model trained on the larger area, which is six times the size from the previous work, for both 8-bit Landsat 5/7, and 16-bit Landsat 8 sensors; and generation of the network architecture to produce LULC maps for the unlabelled 12-bit Sentinel-2 data with the knowledge extracted from the labelled Landsat 8 data.Manitoba Hydro; MitacsMaster of Science in Applied Computer Scienc
Crystal structure and computational study of an oxo-bridged bis-titanium(III) complex
Get PDFThe solid-state structure of the new compound mu-oxido-bis[dichloridotris(tetrahydrofuran-kappa-O)titanium(III)], [Ti2Cl4O(C4H8O)6], at 150 K has been determined. The crystal has monoclinic (C2/c) symmetry and the complex features C2
symmetry about the bridging O atom. Positional disorder is evident in one of the three tetrahydrofuran environments. A post-Hartree–Fock computational analysis indicates that the complex has nearly degenerate triplet and singlet spin states, with the former favoured slightly by ca 2 kJ mol-1.Funding for this research was provided by: Natural Sciences and Engineering Research Council of Canada (grant No.RGPIN-2019-06725).https://doi.org/10.1107/S205322962100609
Hard Infrastructure, Hard Times: Workers Perspectives on Privatization and Contracting out of Manitoba Infrastructure
No full textFor several years, the Manitoba government led by the Progressive Conservatives has been pushing hard to reduce the number of government workers, while transferring work and contracts to the private sector. This report examines the push for privatization and contracting out of design and maintenance of Manitoba’s infrastructure and transportation services. The report focuses on gathering the perspective of government workers who are or were responsible for a variety of tasks such as highway and bridge maintenance, including snow clearing, capital project planning and delivery, road safety and enforcement, including regulation of trucking, maintenance of the provincial vehicle and equipment fleet, operation of water structures and ferries, as well as winter roads. Findings, based on reports from workers, include:
Short-staffing is jeopardizing public safety and leading to burnout.
Workers predicting a reduction in quality of service and assets from the changes.
Workers expecting higher costs for taxpayers and reduced value for money.
Civil service expertise is being ignored, with workers shut out of the process. These negative results regarding service quality and public safety are consistent with earlier studies, with Manitoba ignoring the evidence. The report concludes that the destruction of internal capacity built up over decades will be costly and challenging to undo, and that Manitobans, along with public sector workers, are already paying the price.Social Sciences and Humanities Research Council of Canada, MGEUhttps://mra-mb.ca/publication/hardinfrastructurehardtimes
Effects of multigenerational exposure and phenotypic variation on a freshwater fish species exposed to elevated carbon dioxide (CO2)
Get PDFThe amount of dissolved carbon dioxide (CO2) and the acidity of aquatic ecosystems is increasing as atmospheric CO2 concentrations increase due to human activities. Changes in pH and dissolved CO2 can have considerable aversive effects on fish physiology and behaviour, which can result in negative effects on fish populations. Multigenerational studies have found that the conditions experienced by parents can have significant effects on the performance of their offspring and understanding these effects can help to predict how fish populations will cope in future conditions. Additionally, repeatable behavioural phenotypes are good predictors of trends in behaviour, can be useful predictors of other physiological and life history traits, and can be subject to selection pressures. Unfortunately, the effects of elevated CO2 on freshwater fishes over multiple generations, and the effects of behavioural phenotypes, are poorly understood. In my thesis, freshwater Japanese Medaka (Oryzias latipes) were used to investigate the influence of phenotypic variation and differences in time of exposure (generational) on biological responses to elevated CO2. Lab-reared medaka were divided into ‘responsive’ and ‘non-responsive’ groups based on behavioural differences from the population mean during acute exposure to high CO2 in a common shuttling and novel tank behavioural assay. Responsive and non-responsive fish in parental generation (P) were subdivided and exposed to either control (~480 ppm) or high CO2 (~1250 ppm) conditions over a 6-week period. Following this time, eggs from this generation were collected and randomly selected into either high or control conditions, where they were hatched and reared until maturation (filial generation one (F1), 18 weeks). Eggs from F1 were collected and hatched and reared in the same conditions as their parents until adulthood (filial generation two (F2), 24 weeks). Body condition (size, weight and length), behaviour (total distance moved, time spent in the outer zone of the behavioural arena, and swimming direction), reproductive (number of eggs, size of eggs, and survival to hatch) performance, and the relative abundance of various mRNA transcripts in whole brain tissue of fish was measured across these three generations. Behavioural phenotypes influenced reproduction for P and F2 generation fish, and growth for F1 and F2 fish; suggesting that intraspecific variation in behavioural phenotypes may influence how medaka respond to elevated CO2. However, behavioural phenotypes did not have a significant effect on mRNA abundance on genes targeted in my study. Multigenerational exposure to elevated CO2 were shown to improve the performance of offspring in some measures and resulted in changes of mRNA abundance of several genes. Transgenerational exposure, where a parent or grandparent was exposed to elevated CO2 but the offspring were not exposed to elevated CO2, resulted in some deleterious effects suggesting that, generally, exposure to environmental conditions that differ from that of their parents may put fish especially at risk. In my thesis, current CO2 exposure appeared to be the best predictor of overall condition, where fish exposed to elevated CO2 were worse off than fish exposed to control CO2 conditions. The results of this research contribute to filling a current gap of knowledge in understanding how freshwater fish will respond to future conditions over an ecologically-relevant time scale. Importantly, this information will contribute to generating more informed decisions on freshwater ecosystem management and future research directions. Marine and freshwater environments offer food and water security and are of high importance to the economy and the health of our planet, making my research relevant to our broader society."This research was funded by a Natural Sciences and Engineering Research Council (NSERC) NSERC Discovery Grant, and a Research Manitoba Early Career Researcher Grant held by C. T. Hasler, and I was supported by an NSERC Canada Graduate Scholarship."Master of Bioscience, Technology, and Public Polic
Analysis of Impact of Alcohol on Brain's Activity
Get PDFElectroencephalography is an electrophysiological monitoring process to capture electrical activity on the scalp that has been shown to represent the macroscopic activity of the surface layer of the brain underneath. It is typically non-invasive, with the electrodes placed along the scalp. Computer programs in different programming language such as MATLAB, Python are used to simulate and study brain signals. This thesis focuses on utilizing Python, an open-source programming language to understand the impact of alcohol on one’s memory and attention and comparing them with non-alcoholic brain. To carry out this research, we are using open-source EEG data collected from alcoholic and non-alcoholic subjects subjected to visual stimuli. Experiments are carried out to observe spatial patterns related to both groups' brain activity and their association with different region of brain such as memory, attention, somatosensory, and emotional regulation regions. Besides the spatial pattern, we are also focusing to find source signals and their association with respect to attention region to understand the impact of alcohol on one’s attention function. Finally, the optimal sources based on optimal alpha and gamma rhythms are estimated. For these optimal source channels, we estimated time-frequency based spectrogram to understand the association of other band powers for both groups. Beta power activities from these spectrograms are analyzed for both groups to understand attention-deficit caused by alcohol consumption. By analyzing the results from the experiments can help us understand the impact of alcohol on one's brain's activity.Master of Science in Applied Computer Scienc
