21793 research outputs found
Sort by
Repetitive Financial Ads on Social Media Shape Next-Gen Future Financial Experience: Why Financial Experts Should be Alert?
Get PDFSocial media platforms allow financial institutions and advertisers to reach a broad audience and promote their offerings. Consequently, financial ads have become a common feature in social media feeds. Nevertheless, more knowledge is needed regarding the effects of these ads on the Next Gen customer journey (age and culture), who navigates through the complex financial ecosystem trying to make sense of these repetitive financial ads combined with the interference of social media influencers. This qualitative research explores the impact of repetitive financial advertisements on social media and their influence on the general perception of various age groups that compose the Next-Gen. The methodology consisted of an in-depth content analysis of financial influencers on social media and 15 in-depth interviews of participants from different demographic and culture groups who shared their detailed customer journey experiences, including attitudes and perceptions toward repetitive financial ads. The results of this study contribute to providing a better understanding of a) how exposure to repetitive financial ads shapes the customer’s attitude towards financial products and services, b) how variations in perceptions may depend on the customer’s age groups and cultural background, and c) how social media influencers combined with social media platforms selection may impact the customer’s financial literacy and perception on financial products and services. From a managerial perspective, the power of influencers and social media platforms within financial services should be evaluated beyond
robotic information gathered by big data. Although the qualitative research limits the
aggregate results based on the snowball recruitment and the number of interviewees, it enriches the perspective of putting the human first by listening to the storytelling; it is possible to grasp how the customer engages in sharing, reading, and commenting about financial services and products on various social media platforms, and how the customer is influenced by “financial experts” who promote their expertise on these platforms. Ethical guidance is needed at all levels (e.g., customers, financial institutions, financial advertisers, marketers) to develop and achieve social media efficacy while tailoring educative financial communication strategies toward specific age and cultural groups
Influence of Process Parameters on Microstructural Characteristics and Mechanical Properties of Ti-6Al-4V Samples Manufactured through Electron Beam Melting Manufacturing
Get PDFMetal Additive Manufacturing (MAM) is a manufacturing technique that builds solid components layer by layer. In comparison to traditional manufacturing methods, metal additive manufacturing offers advantages by allowing the creation of intricate geometries while reducing the buy-to-fly ratio of components. Understanding how manufacturing parameters affect the mechanical properties of produced components is applicable to the research and development of materials used in aero-engine and biomedical applications.
The study that is the subject of this thesis examines the microstructure of Ti64 samples printed using Electron Beam Melting (EBM) with specific variations in process parameters, such as adjustments in beam current and scan speed intended to achieve varying microstructural properties. The study compares the variations in process parameters with the types of defects and overall sample porosity. Additionally, this research explores the relationship between process parameters, defect production, and the tensile and fatigue properties of the printed samples.
The results of the microstructure examination indicate that individual process parameters significantly influence print porosity for the same Volumetric Energy Density (VED). However, VED alone does not dictate alpha lath thickness or variations in tensile properties. While porosity and defect size do not notably affect tensile properties, they impact the fatigue life of the samples with larger defects leading to a shorter fatigue life. Optical microscopy analysis of print samples provides a local estimation of defect parameters for an initial assessment and prediction of global porosity in larger printed components
Evaluating Infrastructure Demand and Optimizing Charging Strategies for Battery Electric Bus Fleet - A Pilot Study on Concordia Shuttle Fleet.
Get PDFThe transition from conventional buses to Battery Electric Buses (BEBs) poses significant challenges for transit agencies in terms of feasibility and in identifying potential operational issues. One of the crucial challenges is accurately determining the charging infrastructure demand for effective fleet management of electric buses. Insufficient infrastructure can result in operational problems, increased costs, and dissatisfied passengers. Additionally, high initial and maintenance costs, as well as compatibility issues, further impede infrastructure development. Evaluating infrastructure demand and the performance of different charging strategies in various route and operational conditions is essential in addressing these challenges. This thesis aims to evaluate the charging infrastructure demand and the effect of different charging strategies for a Battery Electric Bus (BEB) fleet using mathematical formulations and simulation modeling, specifically focusing on three scenarios: Depot charging, Depot & Opportunity charging combined, and Opportunity charging. The impact of these scenarios on fleet operations is analyzed using Discrete Event Simulation, with Arena software employed for simulation purposes. Additionally, the thesis evaluates the daily average charging costs, considering appropriate assumptions.
The results of the simulations indicate that both the Depot & Opportunity charging combined and Opportunity charging alone scenarios outperform the depot charging strategy in achieving low charging costs. The analysis ascertained that a battery capacity of 300 kWh, coupled with a charging power of 100 kW, suffices to maintain a 100% trip success rate for the Concordia University shuttle fleet under the route conditions considered. However, it is worth noting that the depot charging strategy with overnight charging takes advantage of lower energy costs and grid loads during non-peak hours with proper charging schedules. Overall, the proposed work provides valuable insights for decision-makers and transit agencies looking to deploy electric shuttle bus fleets across different route conditions
Shifting To Renewable Furan Candidates: Synthesis and Study of New Optoelectronic Materials
Get PDFWith climate concerns consistently growing, the chemical industry must shift away from petrochemicals and turn to sustainable feedstocks derived from lignocellulosic biomass. There is urgency to develop new synthetic methodologies that effectively convert biomass-derived starting materials into useful consumer products. The production of multi-arylated furans and 2,5-furan-based oligomers, has gained popularity as organic candidates for light-emitting diodes or field-effect transistors, owing to their favourable optoelectronic properties found in highly conjugated furan-based materials. Hydroxymethylfurfural and furfural are biomass-derived platform chemicals that have received attention as raw materials due to their chemical functionality, renewability, and capability to produce various furan-containing building blocks. In this work, we aim to develop a synthetic methodology starting from biomass-derived chemicals to design potential optoelectronics. Hydroxymethylfurfural was studied to develop multi-arylated furans; however, a rigid diol monomer was synthesized instead. Working with methyl-5-bromofuran-2-carboxylate led to three different multi-arylated furans via regioselective halogenations followed by Pd-catalyzed cross-coupling reactions. 5-Bromofurfural, a close derivative of biomass-derived furfural, was an ideal candidate to produce ten 2,5-furan-based oligomers and one 2,5-furan-based push-pull chromophore. To rapidly extend the π-conjugation of each material, a double Pd-catalyzed decarboxylative cross-coupling (DCC) reaction was utilized. This strategy fuses two nucleophilic arylated furan acids with a dihalogenated aryl linker to produce highly planar furan-based oligomers. Regarding overall yields of the double DCC, electron-neutral and electron-donating furan acids exhibited higher yields (32 – 74%) than electron-withdrawing acid derivatives (18 – 21%). All furan-based compounds were then characterized by NMR and their absorbance, photoluminescence and quantum yields were studied in dilute solution
The neuroscience of creative perception: From visual ambiguity to neuroacoustics
Get PDFThis thesis advances the understanding of creativity by exploring its connection with sensory perception, particularly through divergent perception, a concept that extends beyond traditional divergent thinking approaches. It posits that creative ideas emerge not only from how we generate thoughts but fundamentally from how we perceive our environment, emphasizing the role of sensory ambiguity in the creative process. By integrating cognitive neuroscience with music theory, this work examines the relationship between sensory ambiguity and creative perception. Using magnetoencephalography (MEG), it investigates the neural underpinnings of divergent perception and its interaction with stimulus properties and individual creativity levels through a series of theoretical work and experimental studies (Chapters 1 to 3). These studies reveal that the brain's processing of ambiguous stimuli and its spontaneous activity are influenced by the individual's level of creativity, particularly when exposed to scale-free visual stimuli. Transitioning from visual to auditory perception, the thesis introduces a musically-inspired model of brain dynamics, emphasizing the creative potential of biosignals and their inherent harmonic complexity for musical exploration (Chapters 4 and 5). This includes the development of a Python toolbox for harmonic analysis of biosignals, designed to facilitate artistic and scientific exploration of the latent harmonies in living systems. Through this interdisciplinary approach, the thesis underscores the significance of divergent perception in creativity and offers new tools and models for investigating the complex interplay between sensory ambiguity, perception, and creative output
A Deep Few-Shot Network for Protein Family Classification
Get PDFProtein sequence analysis is arguably a challenging modern bioinformatics problem covering various applications such as disease research, precision medicine, and therapeutics. Given the emergence of sequencing technologies and the resulting large-scale databases, protein family classification is an open problem in bioinformatics. Recent advances in computer science have opened new gates to researchers in various scientific domains. Bioinformatics, as an intermediary research field, takes advantage of these advancements from conventional machine learning methods to large language models, and biostatistics. Utilized machine learning techniques for protein family classification, are dependent on domain experts to generate features which could be time-consuming and challenging. Deep learning algorithms have shown promising results in proteomics; however, their application is limited to the availability of massive data sets for training. Since the required data comes from experiments, it can be highly complex or incomplete. As an alternative, few-shot models can learn and generalize from a few observations. To address the mentioned limitations, in this research, we designed and implemented a deep few-shot network for protein family classification and our result showed outperformance to state-of-the-art baseline models. To the best of our knowledge, this is the first deep network tailored for primary sequence family classification that can highly perform with a very limited number of observations
A Comparison of IoT Communication Libraries: APIs and Performances
Get PDFThe Internet of Things (IoT) and the number of deployed IoT devices are growing exponentially nowadays. These devices play pivotal roles in diverse domains, e.g., smart homes. Given their constrained processing and memory capacities, IoT devices communicate with one another through specialized protocols.
The two main IoT communication protocols are the Constrained Application Protocol (CoAP) and Message Queuing Telemetry Transport (MQTT). By March 1, 2023, there were 35 public libraries of CoAP and 40 of MQTT. These libraries have different characteristics, including levels of completeness and runtime performances.
Because of diverse requirements in different domains, the same protocol/library does not apply to any applications. Consequently, developers must select a library (e.g., Californium, java-coap, Paho MQTT, or HiveMQ MQTT Client) but they do not have access to comprehensive and clear comparisons of the API and performance of these protocols and their implementations, impeding their ability to make informed choices.
In this thesis, we implement multiple IoT scenarios using the CoAP and MQTT protocols and several of their implementation libraries. We conduct a comprehensive comparative analysis based on API and performance metrics, including static metrics, packet sizes, and runtime performance.
We thus provide developers with evidence to choose between CoAP and MQTT protocols and their libraries. In future work, we will expand this work to include other IoT protocols and libraries, more scenarios and metrics
Modeling and Attenuation of Motion Artifacts in Capacitive ECG
Get PDFTo reduce the burden on the health system and democratize access to health care, current research is aimed at embedding medical devices in daily-life objects with autonomous diagnostic algorithms. In the case of popular tools such as the electrocardiogram (ECG), electroencephalogram (EEG), and electromyography (EMG), one of the scientific challenges is replacing the standard wet Ag/AgCl electrode. A strong candidate is the capacitive electrode, which can be seamlessly integrated into chairs, beds, car seats and wearable devices. This is a dry and active kind of electrode, fabricated on a rigid or flexible printed circuit board. Although in ideal conditions capacitive electrodes can provide high-quality biopotential measurements, they are prone to motion artifacts (MAs) because they do not stick to the patient's body. A MA is a large interference that can render the ExG analysis impossible. Often, it is much larger than the targeted signal and it can even saturate the analog front-end's input. MAs are often described as random or unpredictable events, however, in this dissertation they were modeled based on triboelectric nanogenerator theory. The proposed model uses information on displacement and speed to mimic the MA behavior. It also supports existing bibliography that MA comprises two main phenomena, a change in electrode capacitance (capacitance between electrode and patient) and generation of triboelectricity. The electrode capacitance variation can cause voltage division with the input capacitance (reduce signal to noise ratio), low cut-off frequency fluctuation (common-mode signals converted into differential artifacts) and modulation of DC voltages across the electrode capacitance (voltage spikes). To attenuate the effects of electrode capacitance variation, three topologies of capacitive electrode are proposed: i) a through-body negative feedback is applied to stabilize the electrode's gain; ii) the input resistance is boosted with a positive feedback so a series capacitance can be inserted; iii) a control system detects the electrode capacitance change and modifies the input resistance
The Use of Songwriting to Explore Anxiety Related to Food Allergies: A Heuristic Self-Inquiry
Get PDFLife-threatening food allergies affect individuals on various levels, from their quality of life and mental health to the threat they impose on their safety. There is a gap in existing research that examines how music therapy might help individuals address food allergy related anxiety. As someone who has lived with a life-threatening peanut allergy my entire life, I have experienced the many complications and challenges that one faces when it comes to navigating their way through the world. The purpose of this self-heuristic inquiry was to use songwriting as a means to explore my food allergy-related anxiety. I generated data by composing a total of four original songs, each of which explored a specific topic using a writing prompt, regarding my experience of living with a food allergy. Data analysis produced three main categories that provide insight into broad perspectives on my experience, or what I call, the big picture, experiences of the additional labour that living with an allergy requires, and the varied emotions that I experience in diverse contexts. I conclude with a creative synthesis; an original song that synthesizes insights derived from the research process. Finally, this paper concludes with implications for my music therapy practice which include insights about how I can utilize the tools that I have gained throughout this process to support individuals through music therapy who live with life-threatening food allergies
Exploring Convex Optimization and Transformer based Methods for Efficient Visual Object Tracking
Get PDFThe effectiveness of a visual object tracking algorithm heavily relies on how well it represents the target object through a collection of feature templates, or channels. However, some channels lose their discriminative power during challenging video conditions such as target deformation, occlusion, and motion blur, leading to tracking failures. Discriminative Correlation Filter-based (DCF) trackers address these video challenges by aggregating hand-crafted and deep Convolutional Neural Network-based (CNN) channels. However, this approach increases the computational complexity of the tracker and significantly reduces inference speed, especially on constrained hardware such as a Central Processing Unit (CPU) or edge devices. We observe a parallel trend in end-to-end trainable deep Siamese Network-based (SN) trackers, which deploy parameter-heavy backbones for feature extraction and rely on specialized hardware such as a Graphics Processing Unit (GPU) for faster inference. In this thesis, we propose computationally efficient solutions to both DCF and SN tracking algorithms while improving their accuracy.
For multi-channel DCF tracking, we present three solutions to alleviate the impact of non-discriminative features (or channels). These methods leverage the concept of reliability to quantify the discriminative power of a feature (or a channel) based on its filter response. The proposed solutions dynamically lower the weightage of unreliable features (or channels) while emphasizing the temporal smoothness of the learned weights. We formulate the process of learning adaptive weights as a convex optimization problem and derive efficient solutions to maintain tracking speed. Expanding on the lightweight SN tracking paradigm, our first algorithm, MVT, employs a cascaded arrangement of CNN and transformer blocks in its backbone. This approach fuses template and search regions during feature extraction to generate superior feature encoding for target localization. Our second tracking algorithm, a Separable Self and Mixed Attention Transformer-based tracker (SMAT), further increases the efficiency of MVT by replacing the standard attention with a computationally efficient separable attention block. Proposed trackers exhibit superior performance on eight challenging benchmarks compared to the related lightweight trackers, with SMAT emerging as the top performer. The computationally efficient architecture enables our MVT and SMAT trackers to run at real-time tracking speed on a CPU, while achieving a high speed of 150 frames-per-second on a GPU