Heriot-Watt University
ROS: The Research Output Service. Heriot-Watt University EdinburghNot a member yet
4689 research outputs found
Sort by
An investigation into online shopping behaviour in the Canadian apparel sector : recommendations for retailer online capitalisation
Get PDFOnline shopping has distinguished its presence in Canada through attractive prices,
convenience, and a plethora of product choices. More so, whilst rapid advancements in
technology have provided numerous benefits that thrust online shopping technology
forward, they have also led to an increase in customer expectations in terms of the type
of service expected from online retailers. The rise of social media alone has given new
meaning to the concept of information sharing, forcing a degree of transparency now
required by online retailers and expected by shoppers. Consequently, businesses
operating solely on a brick-and-mortar basis may soon find themselves left behind as
the market continues to demand not only an online presence, but one that meets the
already high customer expectations.
In an effort to assist Canadian retailers with creating or enhancing their online presence,
this study looks to investigate online shopping in Canada, and particularly within the
apparel sector. This was done through a number of stages. An examination of past
online shopping research was first conducted in order to extract the most relevant
factors that significantly influence the intention to shop online. These were then
integrated with the Technology Acceptance Model (TAM) in order to develop a
research model that can explain online shopping intention. The model was tested
through a transformative paradigm through which semi-structured interviews were first
conducted through a qualitative study, after which the results were used to inform a
quantitative study questionnaire distributed on social media to gather views of online
shopping.
The data collected from the questionnaire was analysed using structural equation
modelling (SEM) and sought to evaluate the significance of the various relationships
presented in the research model. The results indicated that all but two relationships in
the model were significant. The first relationship pertained to that between information
gathering and perceived usefulness, and the second relationship concerned the effect of
web design as a moderating variable on the former relationship. The research model
highlighted the importance of key constructs related to online shopping intention,
namely, information gathering, experience, trust, enjoyment, perceived ease of use,
perceived usefulness, web design, and attitude. A particular emphasis was placed on
web design as it forms the core of any online store
Sensemaking in middle management : disruptive technology change in a car sales business
No full textAbstract and full text unavailable. Restricted access until 01.01.2025. Please refer to PDF
Designing coherent and engaging open-domain conversational AI systems
Get PDFDesigning conversational AI systems able to engage in open-domain ‘social’ conversation is extremely challenging and a frontier of current research. Such systems are
required to have extensive awareness of the dialogue context and world knowledge,
the user intents and interests, requiring more complicated language understanding, dialogue management, and state and topic tracking mechanisms compared to
traditional task-oriented dialogue systems. Given the wide coverage of topics in
open-domain dialogue, the conversation can span multiple turns where a number of
complex linguistic phenomena (e.g. ellipsis and anaphora) are present and should
be resolved for the system to be contextually aware. Such systems also need to be
engaging, keeping the users’ interest over long conversations. These are only some
of the challenges that open-domain dialogue systems face. Therefore this thesis
focuses on designing dialogue systems able to hold extensive open-domain conversations in a coherent, engaging, and appropriate manner over multiple turns.
First, different types of dialogue systems architecture and design decisions
are discussed for social open-domain conversations, along with relevant evaluation
metrics. A modular architecture for ensemble-based conversational systems is
presented, called Alana, a finalist in the Amazon Alexa Prize Challenge in 2017 and
2018, able to tackle many of the challenges for open-domain social conversation.
The system combines different features such as topic tracking, contextual Natural
Language understanding, entity linking, user modelling, information retrieval, and
response ranking, using a rich representation of dialogue state.
The thesis next analyses the performance of the 2017 system and describes the
upgrades developed for the 2018 system. This leads to an analysis and comparison
of the real-user data collected in both years with different system configurations,
allowing assessment of the impact of different design decisions and modules.
Finally, Alana was integrated into an embodied robotic platform and enhanced
with the ability to also perform tasks. This system was deployed and evaluated
in a shopping mall in Finland. Further analysis of the added embodiment is presented and discussed, as well as the challenges of translating open-domain dialogue
systems into other languages. Data analysis of the collected real-user data shows
the importance of a variety of features developed and decisions made in the design
of the Alana system
Non-linear partial differential equations of kinetic type
Get PDFThis thesis is concerned with the analytical study of non-linear partial differential equations (PDEs) of kinetic type which admit multiple stationary solutions. We consider
a kinetic model which is given by a non-linear PDE in the sense of McKean and describes the time-evolution of the density ft = ft(x, v), (x, v) 2 T ⇥ R, of a collection
of interacting particles moving in the one dimensional torus. We focus on tackling the
main diculties which arise from the fact that the non-linear PDE has unbounded coecients, is non-elliptic and not in gradient form. In particular, we employ techniques
to show well-posedness of the solution ft in a weighted Lp space. When the density ft
does not depend on the spatial variable, we study the long time behavior of the spacehomogeneous PDE by using and comparing two different approaches: hypocoercivity
theory and gradient flow theory.Engineering and Physical Sciences Research Council (grant EP/L016508/01
Reasoning and understanding grasp affordances for robot manipulation
Get PDFThis doctoral research focuses on developing new methods that enable an artificial agent
to grasp and manipulate objects autonomously. More specifically, we are using the concept
of affordances to learn and generalise robot grasping and manipulation techniques. [75] defined affordances as the ability of an agent to perform a certain action with an object in a
given environment. In robotics, affordances defines the possibility of an agent to perform
actions with an object. Therefore, by understanding the relation between actions, objects
and the effect of these actions, the agent understands the task at hand, providing the robot
with the potential to bridge perception to action. The significance of affordances in robotics
has been studied from varied perspectives, such as psychology and cognitive sciences.
Many efforts have been made to pragmatically employ the concept of affordances as it
provides the potential for an artificial agent to perform tasks autonomously. We start by reviewing and finding common ground amongst different strategies that use affordances for
robotic tasks. We build on the identified grounds to provide guidance on including the concept of affordances as a medium to boost autonomy for an artificial agent. To this end, we
outline common design choices to build an affordance relation; and their implications on
the generalisation capabilities of the agent when facing previously unseen scenarios. Based
on our exhaustive review, we conclude that prior research on object affordance detection
is effective, however, among others, it has the following technical gaps: (i) the methods are
limited to a single object ↔ affordance hypothesis, and (ii) they cannot guarantee task completion or any level of performance for the manipulation task alone nor (iii) in collaboration
with other agents. In this research thesis, we propose solutions to these technical challenges.
In an incremental fashion, we start by addressing the limited generalisation capabilities
of, at the time state-of-the-art methods, by strengthening the perception to action connection through the construction of an Knowledge Base (KB). We then leverage the information
encapsulated in the KB to design and implement a reasoning and understanding method
based on statistical relational leaner (SRL) that allows us to cope with uncertainty in testing
environments, and thus, improve generalisation capabilities in affordance-aware manipulation tasks. The KB in conjunctions with our SRL are the base for our designed solutions
that guarantee task completion when the robot is performing a task alone as well as when in
collaboration with other agents. We finally expose and discuss a range of interesting avenues
that have the potential to thrive the capabilities of a robotic agent through the use of the
concept of affordances for manipulation tasks. A summary of the contributions of this thesis
can be found at: https://bit.ly/grasp_affordance_reasonin
Who uses it and who loses it? Personality, activity engagement and cognitive health in old age
Get PDFIdentifying strategies to promote cognitive health in older age is a key research priority
as older adults continue to make up a growing proportion of the global population. The
‘use it or lose it’ theory proposes that leading a more active, engaged lifestyle can be
cognitively protective. Cross-sectional studies provide support for this, with evidence
suggesting that older adults who are mentally, physically, socially and creatively active
in their everyday lives may also have higher levels of cognitive ability and experience
lower levels of cognitive decline. Intervention studies can provide further insight, going
beyond cross-sectional associations to explore causality by testing the effect of
increased engagement in an experimental paradigm. It is essential that such
interventions consider the importance of individual differences; in particular, evidence
suggests that individual differences in personality might predict activity engagement,
and in turn cognitive health. It is therefore possible that individual differences in
personality might influence engagement level within an intervention, and in turn the
degree of benefit received. The PhD research reported in the present thesis examined
these possibilities using data collected from a large-scale, activity-based intervention
study known as The Intervention Factory. This study tested the cognitive benefits of
activity engagement in a more real-world environment by using existing, community based classes and groups. A sample of 336 adults aged 65 and over without any
diagnosed cognitive impairments were recruited and completed baseline assessments.
Cross-sectional data at baseline were used to examine whether lifestyle variables such
as activity engagement mediated any associations between Big Five personality traits
and cognitive ability across several domains. Higher Openness to Experience and lower
Neuroticism and Extraversion predicted higher levels of cognitive performance, but
there was no evidence to suggest these associations were mediated by activity
engagement. The PhD research then examined whether personality might influence
activity engagement and cognitive change within the context of an intervention. A
systematic review of the literature found ten studies that had previously explored this
question; there was some evidence that higher Openness to Experience was linked to
greater cognitive gains when studies used novel intervention methods. This theory was
then tested within the context of The Intervention Factory specifically. Participants were
pseudo-randomly allocated to one of five activity groups (computer classes,
dance/exercise/sport classes, social/bingo groups, language classes or
handicraft/woodcraft classes) or a no-contact control group and attended their activity
for around ten weeks. None of the activity groups showed evidence of significantly
greater cognitive improvements compared to the control group over the course of the
intervention. There was also no reliable evidence that individual personality traits
predicted adherence or moderated intervention-related cognitive change. While these
results did not support the efficacy of real-world activities to promote cognitive health,
several challenges were identified that will inform and encourage future research in this
area. These challenges included issues arising from non-random group allocation,
difficulty recruiting an effective control group and variability in intervention delivery
when translated to a more real-world setting. Addressing these challenges in future
studies will provide further opportunities to explore the potential cognitive benefits of
real world activities, and whether any benefits vary at the individual level
Age heaping in population data of emerging countries
Get PDFMortality analyses have commonly focused on countries represented in the Human
Mortality Database that have good quality mortality data. In this thesis, we address
the challenge that, in many countries, population and deaths data can be somewhat
unreliable. In many countries, for example, there is significant misreporting of age
in both census and deaths data: referred to as “age heaping”. The purpose of our
research is to develop Bayesian computational methods for fitting a new model for
misreporting of age for countries where their population data and death counts have
been affected by age heaping. The innovation of our model is that it allows us
to detect misreporting, identify age preferences and estimate the true underlying
distribution of ages
Synthesis and functionalisation of various magnetic nanoparticles for enzyme immobilisation and cascade reactions
Get PDFEnzymes can be considered as sustainable catalysts due to their high activities with lower
energy requirements while being derived from renewable resources. Enzymes have
shown high chemo-, regio- and enantioselectivity which allows for higher conversion
efficiencies and reduction in waste generation. However, they can be sensitive to the
reaction environment and can denature when used beyond their normal operating
conditions. To improve their operational stability, immobilisation of enzymes on a solid
support can be adapted. This would also allow for the possibility of continuous
processing, ease of separation, recyclability and recovery. Magnetic nanoparticles
(MNPs) provide an elegant solution for enzyme immobilisation because of their high
surface area to volume ratio, enhanced separation/recovery using an external magnetic
field and easily modifiable surface.
This thesis investigates enzyme immobilisation using Fe3O4 MNPs through covalent
binding and NiFe2O4 MNPs via his-tag immobilisation. Initially a model protein and
enzyme are immobilised onto a (3-aminopropyl)triethoxysilane (APTES) functionalised
Fe3O4 MNPs by covalently binding the protein/enzyme to the available amine group.
Then using NiFe2O4 MNPs, two enzymes halomonas elongata omega transaminase
(HeωT) and D-phenylglycine transaminase (D-phgAT) are immobilised through their
available his-tags. The application of NiFe2O4 MNPs was taken further and used in a
one pot purification and immobilisation of HeωT, D-phgAT and Bacillus subtilis glucose
dehydrogenase (Bs-GDH) directly from cell lysates. Finally, a free and immobilised tri-enzymatic system for the production of a 1-phenylethanol from s-methylbenzyl amine
using HeωT, GDH and alcohol dehydrogenase (ADH) from saccharomyces cerevisiae is
studied. Generally, the studies carried out are shown to be successful and highlight the
versatility of MNPs for enzyme immobilisation especially how easily they can be
separated, recycled and recovered.Engineering and Physical Sciences Research
Council (EPSRC) CRITICAT CDT fundin
A study of epidermal and scratch wound healing using a non-local mathematical model
Get PDFThis thesis investigates the impact of cell adhesion-induced adhesion movement
using partial integrodifferential equations. To apply cellular adhesion to biological
phenomena, we are concerned about the healing of epidermal wounds and scratch
wound healing assays. These form two main parts of our study.
We begin by showing how the healing of epidermal wounds is modelled in a
continuum approach. Two cellular mechanisms essentially govern the closure of an
epidermal wound: one is cell proliferation at the wound edge, and the other is cell
migration towards the wound centre. Although many biological factors play a role
in managing these dynamics, the keratinocyte growth factor (KGF) is one of the
biological structures observed immediately after injury and acts as an agent. We
present a study that models how KGF directs cell proliferation and its effectiveness
in cell movement. This study also explores the effect of additional KGF on the
re-epithelialisation rate.
Local continuum models consist of equations with diffusion terms describing
the random motion of cells and reaction terms detecting their kinetic behaviour.
However, cells not only show random motion during their migration towards their
cellular behavioural stimuli; they also move passively due to the attractive forces
arising during their interaction. And this passive movement is called the adhesion
movement. Therefore, we extend our continuous model from our first study by
adding a term corresponding to adhesion motion. This approach sheds light on how
the adhesive structure responds to wound healing compared to epidermal healing
models in which this dynamic is omitted.
Some attractive and repulsive forces occur between cells due to the adhesive
molecules providing interaction between cells. As cells interact with their peers in
a community of different cells, they bind with their particular adhesion molecules,
leading to different-magnitude forces. To support this consequence, we give a model
development in a different perspective. Our non-local model includes two different
adhesion terms depending on the mechanisms of the two most popular adhesion
molecules, E-cadherin and N-cadherin. We suggest that each model corresponds to
the biological structure of E-cadherin and N-cadherin, and then we exemplify the
aggregation behaviour of cells based on their functionality.
Our modelling angle is supported by experimental data in the final stage. This
study consists of parameter estimations based on adhesive structure. We solve
the model numerically according to the diffusion coefficient corresponding to each
adhesion strength. We show how well the numerical results corresponding to the
closure rate of the scratched wound agree with experimental data.
We end the thesis with a short conclusion, including the chapter’s primary results
and possible future challenges
Development of innovative xanthan gum – recycled gypsum binder for new generation of sustainable geotechnical materials
Get PDFToday, one of the major challenges for the construction sector is sustainability. The use
of traditional binders such as Portland cement and lime are associated with significantly
high energy consumption and CO2 emissions. Of equal importance, there is an
increasing need for re-using industrial by-products and waste; helping to reduce landfill
disposal rates and, at the same time, to decrease the use of virgin raw materials.
The main objective of this research project was to develop and assess the feasibility of a
novel binder consisting of a combination of xanthan gum and recycled gypsum to use in
ground improvement and construction materials applications. In particular, the study
focused on the effect of treated sand with different mixtures of the two additives in
different conditions on the unconfined compressive strength and water absorption of the
final material. Recycled gypsum was used, in both its dihydrate and hemihydrate forms,
either in its natural state or after treatment with stearic acid. In total, 45 mix
combinations were carefully assessed and investigated. Binder contents varied between
5-15% of the total sample mass, while initial water content ranged from 15-25%.
Further, the impact of varying the production thermal processing conditions of recycled
gypsum hemihydrate, as well as that of specimen preparation techniques, namely
mixing methods and duration, on binder effectiveness were investigated. X-Ray
diffractometry was adopted for the identification of the mineralogical composition of
the gypsum powder, before and after treatment at each of the studied time and
temperature conditions. Additionally, FT-IR analyses and the Sessile drop technique
were employed to identify potential interactions between xanthan gum and both forms
of recycled gypsum, and the hydrophobicity of the stearic acid treated gypsum,
respectively.
Research findings indicate that a xanthan gum –recycled gypsum binder can potentially
be adopted for a series of geotechnical applications. Optimal performance is achieved
from the combination of xanthan gum and recycled gypsum hemihydrate, although only
under specific conditions. As the dominant performance affecting factors were
identified the initial water and binder content together with the xanthan gum to recycled
gypsum ratio, as well as, the adopted mixing time and method for the preparation of the
blends. Despite the recorded increase in gypsum’s hydrophobicity, treatment with
stearic acid does not provide any significant benefits in terms of unconfined
compressive strength and water absorption.Engineering and Physical Sciences Research Council (EPSRC
Access Repository Dashboard
Do you manage ROS: The Research Output Service. Heriot-Watt University Edinburgh ? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!