Theses, Dissertations, and other Required Graduate Degree Essays

Receive updates for this collection

This collection contains digitized SFU theses except for those theses submitted within the last 12 months. If you cannot find the thesis you are looking for please search Recently Submitted Theses as it may be a recently submitted thesis and thus not yet available in Summit.

Design and in pandemic validation of correlation visualisation for sleep data analytics

Author: 
Date created: 
2021-06-17
Abstract: 

Sleep plays an important role in the overall health and well-being of a child. The relationship between sleep and daytime behaviours of children with sleep disorders is understood poorly; different aspects of a child’s routine may interact with each other to contribute to sleep disorders. To diagnose, monitor and successfully treat many medical conditions pertaining to sleep, it becomes imperative to analyse the many aspects of a child’s daytime and sleep behaviours. We built a visual analytic tool for studying the correlation between different variables pertaining to the daily life of the child. The tool allows clinicians to explore how the different aspects of a child’s behaviour and activities affect their sleep and overall well-being. This tool is developed as an extension of an existing tool SWAPP, which allows caregivers and clinicians to log and monitor the child’s everyday data. Later, we performed a remote usability study on the tool to demonstrate the efficacy of the tool. Finally, we generated actionable guidelines for improving the tool from the results of the study.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Chris Shaw
Department: 
Communication, Art & Technology: School of Interactive Arts and Technology
Thesis type: 
(Thesis) M.Sc.

Blanket Creek Provincial Park weedy field restoration plan

Author: 
Date created: 
2021-05-21
Abstract: 

A key management concern for provincial parks is the establishment of invasive species due to their impacts on native biodiversity. Within Blanket Creek Provincial Park there is a 0.24 ha heavily invaded field dominated by hawkweed species and spotted knapweed which developed after a series of natural and anthropogenic disturbances. Restoration actions are required to renew the ecological process of natural succession and shift the vegetation community from its current state to one dominated by native species. The aim of this project was to determine the current site conditions which will inform a restoration plan for the site and act as baseline conditions for future monitoring. This site assessment focused on the characterization of the vegetation and soil conditions. Restoration recommendations focus on promoting the development of a deciduous forest characteristic of the Interior Cedar-Hemlock biogeoclimatic ecosystem classification zone. The restoration recommendations include invasive species management, decompaction, fertilization, mulching, and the planting of native trees and shrubs.

Document type: 
Graduating extended essay / Research project
File(s): 
Supervisor(s): 
Ruth Joy
Department: 
Environment: Ecological Restoration
Thesis type: 
(Project) M.Sc.

United States bank migrations and deposit dollar concentrations

Author: 
Date created: 
2021-06-21
Abstract: 

This thesis incorporates four studies of the geography of bank offices and deposits in the United States (US). The research examines changes in retail bank branch proximity in neighborhoods, state banking law's role in motivating banks to relocate home offices, and the significance of tax avoidance driving deposits’ relocations. Chapter 1 introduces the framework and research questions that emerged from visually exploring geo-spatial banking data. Chapter 2 considers retail bank proximity changes in neighborhoods, classified by income, in urban Florida. It compares three pre- and post-financial crisis bank branch distributions: those merged with government assistance, those that merged unassisted, and those that did not merge. Did the branching decisions made by any of these bank groups disproportionately affect neighborhoods’ proximity to banks? Kruskal–Wallis and post hoc tests suggest that merged banks, which reduced total branches, did not disproportionately impact any neighborhood group. Statistically significant evidence suggests that unmerged banks, which increased total branches, disproportionately improved proximity to high-income areas, filling a spatial void created by closed offices of merged banks. The results suggest that banking regulators indirectly financed the rearrangement of banking offices, conflicting with federal policies aimed at maintaining bank offices near low-income neighborhoods. Chapter 3 examines the forces that drove a massive accumulation of deposits in Delaware and South Dakota, illuminating changes in banking regulation that lured banks from faraway places. Delaware and South Dakota broke longstanding public policy norms by creating bank-friendly regulation of three banking businesses: credit, insurance, and trusts, becoming a preferred legal “home office” for banks seeking regulatory relief. Chapter 4 traces the laws that helped induce Wall Street banks and other commercial firms to migrate to Utah. Utah expanded the scope of a historical anomaly in US banking regulation, the Industrial Loan Bank, which is exempt from longstanding regulatory norms separating banking from a non-banking business. The final chapter considers the lopsided share of deposits in Delaware, Nevada, South Dakota, and Utah after the flight of deposits from high-tax states. This research contributes to and suggests research possibilities on the oft-neglected subject of fiscal geography.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Geoff Mann
Department: 
Environment: Department of Geography
Thesis type: 
(Thesis) Ph.D.

Towards a better future: How Engage Books creates books that make a difference

Author: 
Date created: 
2021-04-27
Abstract: 

This report looks at the changing landscape of Engage Books as they switch their focus from publishing classic titles to publishing children’s books under the mandate ‘books that make a difference,’ and the tactics they are implementing to push boundaries within the children’s publishing industry. To provide context as to where Engage Books stands as an independent children’s publisher, the report gives a brief overview of the history of the acceptability of sensitive topics in children's literature and the relationship between censorship and small presses. Engage Books has adopted the philosophy that it is easier to shape the minds of children than it is to change the minds of those who are already set in their ways, and thus, has begun introducing previously censored information and major world crises to children in an attempt to help the next generation become informed and engaged citizens who can help create a better society.

Document type: 
Graduating extended essay / Research project
File(s): 
Supervisor(s): 
Scott Steedman
Department: 
Communication, Art & Technology: Publishing Program
Thesis type: 
(Project) M.Pub.

Experimental methodology and its applications in economics

Author: 
Date created: 
2021-03-24
Abstract: 

This dissertation explores and applies experimental methods in economics. The first two chapters deal with the methodology of lab experiments, while the third presents a study on mobility apps. In the first chapter, I examine deliberating groups in a jury-like setting where subjects have private information and an opportunity to discuss it before a vote. The study uses a belief elicitation mechanism to incentivize subjects to truthfully report their beliefs both before and after they deliberate, allowing for the measurement of the change in beliefs. I find that deliberation tends to reduce the average error in beliefs, measured as the difference between the belief and the true outcome. The basic experiment follows past deliberation experiments in the literature. It features an abstract setting with private signals in the form of a randomly drawn red or blue ball. To test whether the results are generalizable, I replicated this experiment in a framed setting where subjects read the evidence from a real murder trial. I found no difference between the results of the experiments in these two different settings. The second chapter investigates the use of reinforcement methods in lab experiment instructions. We experimentally compare how methods of delivering and reinforcing experiment instructions impact subjects' comprehension and retention of payoff-relevant information. We find combinations of reinforcement methods that can eliminate half of non money-maximizing behaviour, and we find that we can induce a similar reduction via enhancements to the content of instructions. Residual non money-maximizing behaviour suggests this may be an important source of noise in experimental studies. The third chapter diverges from lab experiments to study Mobility as a Service (MaaS). We test whether a multimodal route-planning service caused users to use combined routes featuring both ride hailing and transit. We find that ride-hailing trips connected with rail stops increased from 3.0% of trips to 5.5% among existing users. In areas where the feature supported bus connections, trips connecting to bus stops increased from 4.6% to 8.7% among existing users.

Document type: 
Thesis
File(s): 
Supervisor(s): 
David Freeman
Department: 
Arts & Social Sciences: Department of Economics
Thesis type: 
(Thesis) Ph.D.

Automating data preparation with statistical analysis

Author: 
Date created: 
2021-03-18
Abstract: 

Data preparation is the process of transforming raw data into a clean and consumable format. It is widely known as the bottleneck to extract value and insights from data, due to the number of possible tasks in the pipeline and factors that can largely affect the results, such as human expertise, application scenarios, and solution methodology. Researchers and practitioners devised a great variety of techniques and tools over the decades, while many of them still place a significant burden on human’s side to configure the suitable input rules and parameters. In this thesis, with the goal of reducing human manual effort, we explore using the power of statistical analysis techniques to automate three subtasks in the data preparation pipeline: data enrichment, error detection, and entity matching. Statistical analysis is the process of discovering underlying patterns and trends from data and deducing properties of an underlying distribution of probability from a sample, for example, testing hypotheses and deriving estimates. We first discuss CrawlEnrich, which automatically figures out the queries for data enrichment via web API data, by estimating the potential benefit of issuing a certain query. Then we study how to derive reusable error detection configuration rules from a web table corpus, so that end-users get results with no efforts. Finally, we introduce AutoML-EM, aiming to automate the entity matching model development process. Entity matching is to find the identical entities in real-world. Our work provides powerful angles to automate the process of various data preparation steps, and we conclude this thesis by discussing future directions.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Jiannan Wang
Department: 
Applied Sciences: School of Computing Science
Thesis type: 
(Thesis) Ph.D.

Bakhtin’s philosophy for education: Pedagogy of aesthetics and dialogism

Author: 
Date created: 
2021-03-08
Abstract: 

This thesis is an attempt to analyze the philosophical ideas of Mikhail Bakhtin from the point of view of their possible implementation in educational practices. The first chapter examines the philosophy of Bakhtin's aesthetics, which postulate that the perception of the surrounding world is possible through non-rational methods of phenomenological and hermeneutic reflections. Particular attention is paid to Bakhtin’s understanding of the process of cognition, where a human is presented in his/her holistic subjective being. According to Bakhtin, this idea permits the usage of subjective cognitive processes in objective reality only through an individually responsible act. The second chapter deals with the possible implementation of the Bakhtinan dialogic philosophy in the theory of knowledge. An attempt is made to trace the understanding of the essence of dialogism from the phenomenological premise that any consciousness is a text that includes the cognizer in a situation of aesthetic understanding and, as the result of such participatory thinking, triggers the mechanism of an internal dialogue. The balancing correlation between the idea of “dialogic” and the idea of “answerability” as part of the general life experience is also explored. The third chapter examines Bakhtin’s architectonics as a specific strategic tool, which allows for optimizing and effectively carrying out the many interrelationships of the participating in many educational processes. Additionally, Bakhtin’s concepts of chronotope and nonalibi in existence are considered as categories of theoretical cognition and phenomena that can help better comprehend the truth, especially in post-modern educational conditions. Together with Bakhtin's ideas, the ideas of cognition specific to the neo-Kantian, primarily German, philosophical schools, are explored as well as their influence on Bakhtin and his followers. I also include some memories from my own teaching experience, which I collected for over 25 years of teaching foreign languages from K-12 to the university level.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Natalia Gajdamaschko
Department: 
Education: Faculty of Education
Thesis type: 
(Thesis) Ph.D.

An efficient approach to pruning regression trees using a modified Bayesian information criterion

Author: 
Date created: 
2021-04-14
Abstract: 

By identifying relationships between regression tree construction and change-point detection, we show that it is possible to prune a regression tree efficiently using properly modified information criteria. We prove that one of the proposed pruning approaches that uses a modified Bayesian information criterion consistently recovers the true tree structure provided that the true regression function can be represented as a subtree of a full tree. In practice, we obtain simplified trees that can have prediction accuracy comparable to trees obtained using standard cost-complexity pruning. We briefly discuss an extension to random forests that prunes trees adaptively in order to prevent excessive variance, building upon the work of other authors.

Document type: 
Graduating extended essay / Research project
File(s): 
Supervisor(s): 
Thomas Loughin
Department: 
Science: Department of Statistics and Actuarial Science
Thesis type: 
(Project) M.Sc.

Multilingual unsupervised word alignment models and their application

Date created: 
2021-03-05
Abstract: 

Word alignment is an essential task in natural language processing because of its critical role in training statistical machine translation (SMT) models, error analysis for neural machine translation (NMT), building bilingual lexicon, and annotation transfer. In this thesis, we explore models for word alignment, how they can be extended to incorporate linguistically-motivated alignment types, and how they can be neuralized in an end-to-end fashion. In addition to these methodological developments, we apply our word alignment models to cross-lingual part-of-speech projection. First, we present a new probabilistic model for word alignment where word alignments are associated with linguistically-motivated alignment types. We propose a novel task of joint prediction of word alignment and alignment types and propose novel semi-supervised learning algorithms for this task. We also solve a sub-task of predicting the alignment type given an aligned word pair. The proposed joint generative models (alignment-type-enhanced models) significantly outperform the models without alignment types in terms of word alignment and translation quality. Next, we present an unsupervised neural Hidden Markov Model for word alignment, where emission and transition probabilities are modeled using neural networks. The model is simpler in structure, allows for seamless integration of additional context, and can be used in an end-to-end neural network. Finally, we tackle the part-of-speech tagging task for the zero-resource scenario where no part-of-speech (POS) annotated training data is available. We present a cross-lingual projection approach where neural HMM aligners are used to obtain high quality word alignments between resource-poor and resource-rich languages. Moreover, high quality neural POS taggers are used to provide annotations for the resource-rich language side of the parallel data, as well as to train a tagger on the projected data. Our experimental results on truly low-resource languages show that our methods outperform their corresponding baselines.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Anoop Sarkar
Department: 
Applied Sciences: School of Computing Science
Thesis type: 
(Thesis) Ph.D.

Treatment wetlands for industrial wastewaters: A study of science, policy, and management

Date created: 
2020-07-16
Abstract: 

Bitumen extraction in Alberta’s oil sands region generates large volumes of oil sands process-affected waters (OSPW) that pose environmental and human health risks. Currently, few feasible options for managing these large and growing volumes of polluted waters exist. The primary objective of this research is to investigate the feasibility, effectiveness, and safety of treatment wetlands as a treatment option for the oil sands industry. To do this, a mechanistic model of the fate and toxicity of OSPW contaminants in treatments wetlands was developed and tested in field studies at the Kearl Treatment Wetland – a free water surface flow wetland in northern Alberta. Measuring concentrations of polycyclic aromatic hydrocarbons (PAHs) and naphthenic acids (NAs) in influent and effluent of the Kearl Treatment Wetland showed that the combined total mass of all detected PAHs and NA reduced by 54 to 83% and 7.5 to 69%, respectively, as a result of treatment. Concentrations of PAHs and NAs in the aqueous phase of the wetland were measured using polyethylene (PE) and Polar Organic Chemical Integrative Samplers (POCIS), respectively. The model is shown to be in good agreement with the experimental observations and required only minimal calibration. Application of the model shows that evapotranspiration is not likely to significantly contribute to the removal of OSPW contaminants. Chemical removal relies mainly on transformation in wetland rooting media due to high microbial activity in wetland biofilm. Higher rates of transformation result in greater removal efficiencies for most chemicals. However, highly hydrophobic substances experience low removal efficiencies and appear to be unaffected by changes in transformation rates in the wetland suggesting wetland treatment is not suitable for these substances. Treatment efficiency is sensitive to wetland surface area and flow rate of water through the wetland suggesting intentional wetland design and operation can improve treatment efficiency. Trade-offs in wetland design and operation can be informed by the model.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Frank Gobas
Department: 
Environment: School of Resource and Environmental Management
Thesis type: 
(Thesis) Ph.D.