3855 research outputs found
Sort by
Wood future: forestry, carbon, and wood architecture in Northern Ontario
No full textThis thesis explores the interconnected past, present, and future of forestry and wood architecture in Northern Ontario through a graphic narrative. This narrative is a method of research that explores subjects by reading
and visualizing subject matter and synthesizing it through illustration to
create a narrative alongside a textual literary review. Through this method
of storytelling, it aims to answer the following question: How can Northern
Ontario build upon the history of its forestry to evolve wood architecture
and culture towards a sustainable future? The planet is faced with a climate
crisis. Northern Ontario’s forestry industry holds a key to combat climate
change by sequestering carbon while rejuvenating its communities through
sustainable forest management and wood architecture. This thesis recounts
the two-century long relationship of forests, forest industry, and wood architecture in Northern Ontario to design a new sawmill and fabrication centre.Master of Architecture (M.Arch
North American Society for Sport History (NASSH) Proceedings of the 50th Annual Convention
No full textNorth American Society for Sport History Proceedings Of the 50th Annual Convention, held in Chicago, Illinois Double Tree by Hilton & Virtual via HopIn, May 27-30, 202
Influence and social networks
No full textStudying large-scale social networks can be a complex and challenging task when
considering social media's rapid development. Mapping large networks and studying interactions
present barriers in terms of access to data, limitations on analysis, and an approach to identify
unseen influencers in the network. This study examines how connections between data points
and users in a network can be mapped and understood. This method of mapping connections can
allow a researcher to identify influencers within a network and find optimal routes through which
content can be distributed to a broad group of connected users. This is accomplished by
comparing the role of network groups to that of users. This is done by mapping organizations
and connected groups of students on social media networks over time to identify influential
network members.
The project involves studying several campus and community-based pro-Israel student
groups organized into four geographically themed clusters. Data was collected from Twitter
using various methods, including Python language code and NodeXL. Once the data was
collected and analyzed using network link analysis and statistics for interconnections,
visualizations and sociograms were generated using Gephi. Through analyzing network data for
users and organizations, network statistics and metrics can be calculated to identify network
influencers.
The study shows that otherwise unseen influencers can be mapped within a social
network and that their relative social influence can be identified. Studying organizations and
exponential mapping layers of connected users reveals new connections and patterns. The
relative social influence, position, and communication patterns within a network generate new
insights into network members. Hidden influencers were identified and show a connection
between users and otherwise unknown clusters of the network.
The study results show that influencers can be identified and mapped within large and
complex networks and that their relative social influence can be quantitatively calculated. This
has implications for disseminating information within a network, mapping complex interactions
within a social network, and understanding the structural communication pathways of social
networks. This approach can be used in market analysis, research, and other social networks.Doctor of Philosophy (PhD) in Human Studie
Habitat 2.0: creating gentle density communities in Toronto’s laneways
No full textHow can Toronto’s housing crisis be solved in a way that respects the existing urban fabric of its streetcar suburbs while increasing their population density? How can the affordability crisis be solved with a built-form that is actually affordable, quick to construct, and can be seen by the local community as a positive addition to their neighborhood? This thesis project seeks to explore increasing density in Toronto by further developing the laneway home typology, while cultivating vibrant micro communities within Toronto’s laneways. It will seek to reinvent the laneway dwelling, and Toronto’s laneways as a whole, unlocking their true potential for providing a place to live, work and play. This thesis will look at the history of laneways and laneway housing and its persistence today. It will further explore the contextual issues of affordability in Toronto and explore the possibilities of additional new programming in lanewaysMaster of Architecture (M.Arch
Study of quantum phase transitions and topological phases in chains and ladders
No full textThe ground-state phase diagrams and order parameters of low-dimensional quantum
models are analyzed. Those models in their spin representation are the dimerized
spin-1/2 XY and XYZ chains, and the two-leg ladders with anisotropy and three different dimerization patterns, in the presence of uniform and staggered transverse fields.
The analysis is done by using the effective quadratic fermionic Hamiltonian of models,
resulting from the Hatree-Fock mean-field approximation. In the fermionic representation, those models are equivalent to the generic Kitaev -Majorana chains/ladders with
the proper parametrizations.
An exact solvable model, the XY chain has a rich phase diagram, and its distinct
phases are identified by the local and nonlocal (string) order parameters. We have
calculated all the local order parameters (spontaneous magnetization) and the nonlocal order parameters within the same systematic framework, along with the winding
numbers for all regimes of the phase diagram. By combining the exact and the meanfield methods, the local and string order parameters on the phase diagram of the XYZ
chain, are identified and calculated. We found similar qualitative pictures on the phase
diagrams of XY and XYZ chains, where the corresponding parameters of the latter
model are renormalized by the interaction ∆ = Jz/J. For both models, the topological
nontrivial phase is shown to have a peculiar oscillating order with the period of a four
lattice spacing, not reported before and awaiting for its experimental confirmation.
The detailed analysis of patterns of the string order is given. Moreover, the trivial
phases of both the cases are investigated by local order parameters (components of
spontaneous magnetization).
The special XXZ limit of the model with additional U(1) symmetry is in agreement with the Lieb-Schiltz-Mattis theorem and its extensions, plateaus of magnetization, and some additional conserving quantities. We have shown that within the XYZ
chain, where the plateaus are smeared, the robust oscillating string order parameter
is continuously connected to its XXZ limit. Also, the nontrivial winding number and
zero-energy localized Majorana edge states, as additional attributes of the topological
order, are robust in that phase, even off the line of U(1) symmetry.
The phase diagram of the isotropic two-leg ladder is investigated by calculating
the field-induced magnetization at each point along the external field. In the phase
diagram, the two kind of phases, gapped plateau and gapless Lutinger liquid, (LL) are
identified. In the applied uniform field, those models are in agreement with the quantization conditions of the magnetization plateaus. The existence of the mid-plateau in
the staggered ladder with columnar field, we report for the first time is an indication of
a new spin gapped phase in this type of spin structure. For the staggered and columnar
ladder, the alternating field only modifies the phase boundaries of the phase diagram.
The ladder with the rung dimerization and columnar field exhibits additional quantum phase transition by closing and re-opening the zero-plateau and mid-plateau gapped
phases with respect to the alternating field.
The Hatree-Fock mean-field Hamiltonian of the ladders with an anisotropy and two
dimerization patterns, map onto the sum of two quadratic Majorana Hamiltonians,
which are dual to a sum of two (even/odd) XY quantum chains in the alternating
transverse fields. The mapping between the effective Hamiltonian of the ladder and the
pair of the dual XY chains considerably simplifies calculations of the order parameters,
and analyses of the hidden symmetry breaking. The ground state phase diagram of the
staggered ladder contains nine phases: four of them are conventional antiferromagnets,
while the other five possess the non-local brane orders. Using the dualities and the
newly found exact results for the local and string order parameters of the transverse
XY chains, we were able to find analytically all the magnetizations and the brane order
parameters for the staggered case, as well as the functions of the renormalized couplings
of the effective Hamiltonian. The columnar ladder has three ground-state phases, and
it does not possess a magnetic long-range order. The brane order parameters for these
phases are numerically calculated from the Toeplitz determinants. We expect this
study to motivate the search for the real spin-Peierls anisotropic ladder compounds,
which can undergo the predicted quantum phase transitions with a gap closure and
distinct brane orders.Doctor of Philosophy (PhD) in Material Science
Healing home: exploring the potential of trauma-informed design in foster care group homes
No full textThis thesis explores trauma-informed design’s capacity to improve the health and overall livelihood of youth living in the foster care system. Statistics indicate that many youth in foster
care are living with accumulated trauma. This trauma presents itself through physical and
mental illness, and fosters poor adult life outcomes. Progress in the field of trauma-informed
design and homelessness, has validated the success of this practice in the healing and mitigating of trauma. The collaboration between trauma-informed design and foster care architecture however, has not been studied. In the attempt to improve the life outcomes for this vulnerable population, this thesis proposes a trauma-informed foster care group home. A home
designed in purposeful response to the unique traumas experienced by youth in foster care,
provides a safe residence, and curates a healing environment. This is needed to allow youth in
the system a clearer path forward in becoming happy and healthy adults despite their trauma.Master of Architecture (M.Arch
Trophic ecologies of double-crested cormorants and native piscivorous fishes in Lake Nipissing, Ontario
No full textFisheries assessments have indicated a decline in Lake Nipissing’s walleye (Sander
vitreus) population in recent decades. This has coincided with an increase in doublecrested cormorants (DCCO; Nannopterum auritum) on Lake Nipissing to conspicuous
numbers (3,000 nests in 2012), fueling concerns that cormorant predation may be
adversely affecting walleye recovery. I used carbon and nitrogen stable isotope ratios
(δ13C and δ15N, respectively) to examine the food web structure of Lake Nipissing, and in
particular, the role of DCCO predation and competition in relation to four other native
piscivores; walleye, northern pike (Esox lucius), smallmouth bass (Micropterus
dolomieu) and burbot (Lota lota). Trophic position (TP), estimated from δ15N, and
trophic niche size (SEAc), inferred from the dispersion of individuals in δ13C - δ15N
isotopic space, were used to characterize piscivore niches. MixSIAR stable isotope
mixing models (SIMM) were used to estimate the diet compositions of DCCO and
piscivorous fishes. Among the piscivores, DCCO had the highest reliance on pelagic
resources, the lowest trophic position, and the largest trophic niche size. Piscivorous
fishes had higher levels of trophic niche overlap with each other than with DCCO.
SIMMs predicted that DCCO diet was primarily composed of emerald shiner (Notropis
atherinoides) and logperch (Percina caprodes) with low proportions of all other prey
fishes. Emerald shiner and logperch also dominated the diets of piscivorous fishes
indicating dietary overlap with DCCO to some extent. Juvenile walleye were a relatively
small proportion of the diets of both DCCO and piscivorous fishes based on SIMM
predictions. Currently, there is no indication of limitations in Nipissing’s forage fish prey base so potential for interspecific competition is considered low. The low likelihood of
trophic niche overlap between DCCO and piscivorous fishes and the low contribution of
juvenile walleye to DCCO diet suggests a low impact of DCCO predation on Nipissing’s
recovering walleye population. Stable isotope-based diet inference can complement bioenergetic models and other ecosystem assessment methods to improve our understanding
of how DCCO interact with fish populations.Master of Science (MSc) in Biolog
The influence background fluctuations of electromagnetic fields and biophoton emission has on behaviour: a correlational and experimental investigation
Get PDFNontraditional environmental factors such the geomagnetic field and background biophoton
emission have the potential to influence human brain activity. The relationship between brain
activity and geomagnetic field activity fluctuations as well as background photon emission were
examined over the course of this study. An electroencephalographic database of 184 participants
was amalgamated with a geomagnetic field database and a background photon emission database
to be analyzed. The results showed that increased geomagnetic field activity was positively
correlated with an increased alpha power from the right hemisphere of the brain and that as
background photon emission increased, an increase in theta and alpha activity from the frontal
lobe was observed. We further investigated the effect of geomagnetic fields by observing planaria
behaviour after being exposed to one of six applied electromagnetic fields created to mimic a
geomagnetic storm ranging in intensity from 0.1µT to 3.5µT. Planaria were split into two groups:
control, and acute 10µM nicotine exposure 24 hours prior to behavioural observation. The
behavioural observation results showed that planarian mobility increased when exposed to the
synthetic geomagnetic storm electromagnetic field. Planaria experiencing nicotine withdrawal
exhibited more aversive behaviour after being exposed to any intensity of the synthetic
geomagnetic storm electromagnetic field. The data demonstrates that an electromagnetic field
mimicking a geomagnetic storm can exacerbate aversive behaviour in planaria, especially in
planaria experiencing nicotine withdrawal. In conclusion, both geomagnetic field and
background photon emission correlated with brain activity. This research has given reason to
consider how important non-traditional environmental factors are and how they factor into
day-to-day life for biological individuals.Master of Science (MSc) in Biolog
Social media hate speech detection using explainable AI
Get PDFArtificial Intelligence has invaded various fields in the present times. Be it science, education,
finance, business or social media, Artificial Intelligence has found its applications everywhere.
But currently, AI is limited to only its subset ‘Machine Learning’ and has not even realized its full
potential. In machine learning, in contrast to traditional programming which requires writing
algorithms, it is required to find the algorithm that learns patterns from a given dataset and builds
a predictive model and the computer learns the patterns between input and output based on that.
However, a key impediment of current AI-based systems is that they often lack transparency. The
current AI systems have adopted a black box nature which allows powerful predictions, but these
predictions cannot be explained directly. To gain human trust and increase transparency of AIbased systems, many researchers think that Explainable AI is the way forward. In today’s era, an
enormous part of human communication takes place over digital platforms, for example, through
social media platforms and so does hate speech, which are dangerous for an individual person as
well as the society. These days automated hate speech detection is built on social media platforms
such as Twitter, Facebook, etc. using machine learning approaches. Deep learning models attain a
high performance has low transparency due to complex models, which leads to “trade-off”
between performance and explainability. Explainable Artificial Intelligence (XAI) was used to
create black box approaches interpretable, without giving up on performance. These XAI methods
provide explanations that can be translated by humans without having a depth of knowledge in
deep learning models. XAI characteristics have flexible and multifaceted potential in the hate
speech detection by the deep learning models. XAI thus provides a strong interconnection between
an individual moderator and hate speech detection framework, which is a pivot for the research
study in interactive machine learning. In the case of Twitter, the main tweets are detected for hate speech however retweets and replies are not detected for hate speech as there is no tool to handle
the task to detect the hate speech for in progress conversations. Interpreting and explaining
decisions made by complex AI models to understand the decision-making process of these model
is the aim of this research. While machine learning models are being developed to detect the hate
speech on social media, these models lack the interpretability and transparency on the decisions
made. Traditional machine learning models achieve high performance at the cost of interpretability
and explaining model decisions. The main objectives of this research are, to review and present a
comparison of various techniques used in Explainable Artificial Intelligence (XAI), to present a
novel approach for hate speech classification using Explainable Artificial Intelligence (XAI) and,
to achieve a good trade-off between precision and recall for the method proposed. Explainable AI
models for hate speech detection will help social media moderators and any other users for these
models to not only see but also study and understand how the decisions are made and how the
inputs are mapped to the output. As a part of this research study, two data sets were taken to
demonstrate Hate Speech Detection using Explainable Artificial Intelligence (XAI). Data
preprocessing was performed to remove any bias, clean data of any inconsistencies, clean the text
of the tweets, tokenize, and lemmatize the text, etc. Categorical variables were also simplified in
order to generate a clean dataset for training purposes. Exploratory data analysis was performed
on the data sets to uncover various patterns and insights. Various pre-existent models were applied
to the Google Jigsaw dataset such as Decision Trees, K-Nearest Neighbours, Multinomial Naïve
Bayes, Random Forest, Logistic Regression, and Long Short-Term Memory (LSTM) out of which
LSTM achieved an accuracy of 97.6%, which is an improvement compared to the studies of Risch
et al. (2020). Explainable method like LIME (Local Interpretable Model-Agnostic Explanations)
is applied on HateXplain dataset. Variants of BERT (Bidirectional Encoder Representations from Transformers) model like BERT + ANN (Artificial Neural Networks) and BERT + MLP
(Multilayer Perceptron) were created to achieve a good performance in terms of explainability
using ERASER (Evaluating Rationales and Simple English Reasoning) benchmark by DeYoung
et al. (2019) where in BERT + ANN achieved better performance in terms of explainability as
compared to the study by Mathew et al. (2020).MSc Computational Science
An analysis of lung cancer survival using multi-omics neural networks
No full textA key goal of precision health medicine is to improve cancer prognosis. Despite the fact
that numerous models can forecast differential survival from data, progressive algorithms that can
assemble and select important predictors from progressively complex data inputs are urgently
required. As a result, these models should be capable to provide more information about which
types of data are most significant for improving prediction. Because they are adaptable and account
for data density in a non-linear manner, deep learning-based neural networks may be a feasible
solution for both difficulties. In this study, we use Deep Learning-based networks to get how gene
expression data predicts Cox regression survival in lung cancer. SALMON (Survival Analysis
Learning with Multi-Omics Neural Networks) is an algorithm that collects and simplifies gene
expression data and cancer biomarkers in order to enable prognosis prediction. When more omics
data was comprised in model construction, the results (concordance index = 0.635 and log-rank
test p-value = 0.00881) showed that performance improved. We employ eigengene modules from
the results of gene co-expression network analysis as model inputs in its place of raw gene
expression principles. This algorithm verified specific mRNA-seq co-expression modules and
clinical information, which show crucial roles in lung cancer prognosis, revealing various
biological functions by exploring how each contributes to the hazard ratio. SALMON also
performed well compared to other Deep Learning Survival prognosis models.Master of Science (MSc) in Computational Science