21793 research outputs found
Sort by
An Empirical Study on Learning Models and Data Augmentation for IoT Anomaly Detection
Get PDFThis thesis studies the application and impact of deep learning methods in anomaly detection, a critical area within security applications. While deep learning's popularity is driven by its perceived ability to manage complex patterns in large datasets and perform feature engineering inherently, this thesis questions these assumptions. By revisiting feature selection and data augmentation techniques, this research evaluates their effectiveness in improving the performance of deep-learning-based anomaly detection methods. Furthermore, it examines the impact of other essential factors such as model choice (both traditional machine learning and deep learning), data balancing, and hyperparameter tuning on anomaly detection performance.
From these investigations, the thesis reports that the common beliefs surrounding deep learning are not universally valid, highlighting the need for a framework to evaluate the usefulness of features and data for specific cases. To address this gap, a new framework is proposed, guiding data users and anomaly detection tools toward optimal configurations, including feature selection, model selection, hyperparameters, and data augmentation techniques. The effectiveness of this framework is demonstrated using two major IoT datasets, offering insights into improving anomaly detection systems through strategic and evidence-based approaches
Investigating municipal Access to Information via news coverage of Montreal’s housing crisis
Get PDFThis study seeks to reveal the role of Access to Information (ATI) and public records in local journalism by conducting a thematic analysis of 107 news media articles about Montreal’s housing crisis, and by examining original and previously released Access to Information request packages. This work highlights how local news media have covered the housing crisis thus far, with a deliberate focus on the sources and angles used, and how they might address the issue differently going forward, with increased focus on using official documents and ATI requests in the coverage. Importantly, this study focuses on Montreal’s municipal ATI system, as previous work in academia has mostly focused on either Canada’s federal ATI system or provincial/territorial systems.
This study reveals that, in stories about the housing crisis, local news media have tended to favour the voices of politicians and official statements, while only a few rare outlets sporadically use ATI to deepen their reporting.
This study recognizes that tight deadlines in news work and long delays in the municipal ATI system are in part responsible for local journalists’ heavy reliance on political sources and official statements. However, the result of not using ATI as a journalistic source leads to a journalism that remains at the surface level and fails to provide citizens the information they are entitled to, that would allow them to make more informed decisions about municipal policies related to housing and municipal elections. With this important function of local journalism in mind, this study suggests increased use of ATI requests in gaining a deeper understanding of complex issues that directly target citizens
Paradoxical Value Co-Creation Through Destructive Behaviors: Trash Talking in Dota 2
Get PDFCo-creation is always considered an outcome of positive behaviors. Conversely, co-destruction is mostly treated as a negative outcome of negative or destructive behaviors. However, there are instances that negative or destructive behaviors paradoxically create value. Conducting nine interviews and analysing ten extensive threads, my findings suggest that under circumstances, there are more angles to destructive behaviors rather than being merely destructive! Exploring Dota 2 players’ motifs for trash talking that is a negative behavior, I identified four types of value in trash talking. First trash talking for the pleasure of winning, second trash talking for entertainment and fun, third trash talking for gameplay self-perception enhancement and feeling better about oneself and fourth trash talking for social bonding and friendship ties. These findings contribute to the co-creation and trash talking literatures challenging the predominant ideologies that co-creation is always stemmed in positive behaviors and that trash talking is always a negative behavior
Exploring Songwriting as an Expression of Grief on Cape Breton Island: A Qualitative Interview Study
Get PDFLocated in Nova Scotia, Canada, Cape Breton Island has a rich and diverse musical history, which includes a well known tradition of using songwriting as a mourning practice. In the cultural mosaic of Canada, individuals seeking music therapy bring unique and diverse cultural backgrounds and experiences. Exploring these inherent cultural practices can offer models for integrating songwriting into the bereavement music therapy session while also informing music therapy training programs and future research. Grounded in a social constructivist epistemology (Hillier, 2016) and using qualitative semi-structured interviews, this study explored the use of songwriting as an expression of grief for three Cape Breton Island songwriters. Utilizing inductive thematic analysis (Braun & Clarke, 2006), three themes emerged from the data. The first theme, transformative grief journey, reflects how participants experienced personal growth while moving through a unique journey through their grief. Songwriting helped them process a range of emotional responses, provided a safe outlet for difficult feelings, and supported their ability to move forward. The second theme, cultivating connection, highlights how songwriting reduced feelings of isolation and helped participants feel connected to others through sharing their music. The third theme, intrinsic practice, describes how participants trusted the songwriting process and emphasized its cultural importance as an inherent and meaningful way to express grief
Artificial Intelligence for Spectrum-Aware Autonomous Wireless Networks
Get PDFSpectrum is one of the most vital public resources, carrying wireless communications for mobile phones, satellites, and emergency services. Traditionally, spectrum has been exclusively licensed, resulting in occasional underutilization. However, with the rapid proliferation of wireless devices, spectrum congestion has become inevitable, especially in unlicensed networks such as the Internet of Things (IoT), vehicular, and Unmanned Aerial Vehicles (UAVs), where devices rely on a limited number of public frequency bands that may not support large-scale communications. This contradiction between licensed spectrum underutilization and unlicensed spectrum congestion necessitates a rethinking of spectrum allocation and management strategies. In this thesis, we draw inspiration from Cognitive Radio (CR) technology, which equips radio devices with capabilities such as perception, reasoning, and judgment, and extend it to “intelligent radio” that integrates both cognition and learning capabilities. Our work advocates a shift from traditional model-driven approaches that rely on domain knowledge and strong assumptions to data-driven methods that learn directly from raw data and constant interactions with the environment. With the support of Artificial Intelligence (AI), we design intelligent spectrum borrowing and spectrum-sharing techniques that enable wireless devices to operate opportunistically on licensed bands. Specifically, this thesis explores how AI can endow wireless devices with context-awareness, self-optimization, and self-management capabilities for tasks such as dynamic spectrum access, power management, resource allocation, and ensuring security. Additionally, we develop solutions and frameworks for self-sustaining wireless devices that leverage Energy Harvesting (EH), bringing us closer to the realization of green networks. Our AI-driven algorithms are designed with computational efficiency in mind to minimize the burden on resource-constrained devices.
To drive context-aware intelligence in large-scale cooperative networks, we develop various unsupervised Machine Learning (ML) approaches for spectrum sensing. Unlike existing methods, the proposed frameworks operate without the need for labeled data, prior knowledge of the radio environment, or cooperation between licensed and unlicensed users. The approach ensures robust spectrum sensing while minimizing computational overhead for unlicensed users with limited capabilities. Moreover, we investigate how dimensionality reduction can improve computational efficiency and model generalizability. We expand the use of unsupervised learning to hybrid CR networks to allow devices to detect all licensed network states, opening up new opportunities for dynamic spectrum access.
To improve spectrum reasoning and analysis, we introduce some of the first fully unsupervised, data-efficient deep representation learning frameworks. These frameworks are designed to learn effective and disentangled representations of radio environment data. We demonstrate their effectiveness in significantly enhancing spectrum gap detection in small-scale cooperative networks. Additionally, we tackle key challenges of unsupervised learning, such as sensitivity to initialization and the need for predefined cluster counts. In large-scale networks, we propose a generative deep representation model that not only learns efficient representations but also captures the distribution of radio environment data, enabling the generation of new, unseen samples.
To facilitate edge intelligence and enhance the privacy of intelligent radios, we propose the first fully unsupervised deep Federated Learning (FL) framework for secure and distributed spectrum sensing in large-scale mobile networks. By leveraging user mobility across a large geographical area, the method enhances spatio-temporal diversity without requiring the transmission of private data to a central unit for processing. Instead, data is collected locally, and a shared model is collaboratively trained in a decentralized manner, significantly reducing communication overhead and safeguarding user privacy.
We tackle the growing challenge of spectrum scarcity in Cognitive IoT (CIoT) networks, where the demand for spectrum is increasing due to the expansion of connected devices. To address this, we develop intelligent and adaptive control algorithms for the joint management of network resources in spectrum-sharing environments. First, we formulate optimization problems under various constraints and model the decision-making process of a CIoT agent in the dynamic radio environment. We then propose two novel Deep Reinforcement Learning (DRL) algorithms that enable devices to autonomously learn operational strategies to optimize network resources and maximize long-term throughput without comprehensive prior knowledge. Additionally, we introduce innovative exploration strategies to enhance the CIoT agent's ability to identify optimal actions that maximize data rates. Considering the resource limitations of these networks, the algorithms are designed to be lightweight to reduce computational burdens on users. We also integrate EH techniques, such as Wireless Power Transfer (WPT) and Simultaneous Wireless Information and Power Transfer (SWIPT), to make these networks self-sustaining.
Finally, to develop dynamic strategies for navigating hostile spectrum-sharing environments impacted by jamming attacks, we propose an intelligent DRL approach that does not rely on frequency hopping. This algorithm is designed for rapid convergence, energy efficiency, and adaptability to adversarial conditions. We begin by formulating the optimization problem of power control under various constraints and modeling the decision-making process of the CIoT agent in such a hostile environment. Then, we introduce a novel interference-aware exploration strategy that enables the CIoT device to autonomously learn a transmission strategy, effectively mitigating jamming attacks and maximizing performance. Furthermore, we leverage WPT EH to allow the CIoT agent to convert jamming interference into a valuable resource for recharging.
In summary, the contributions of this thesis lay the foundation for a new generation of intelligent, autonomous wireless networks that are both spectrum-aware and agile, capable of optimizing resources and adapting to dynamic and complex environments
Development of Computation-Efficient Computer Vision Systems for High-Quality Brain Tumor Segmentation
Get PDFIn this thesis, two design methodologies are proposed, and also applied in the development of 2 computer-vision systems for computation-efficient and high-quality brain-tumor detection.
The first methodology aims at developing systems to detect, by conventional image processing procedures, 3D-object locations with a pixel-wise precision. The main operations of the detection are predicting gray-level distribution of the pixels in the object region and, based the prediction result, identifying/removing regions of non-interest. As 3D inputs can be sliced into axial, coronal or sagittal slice series, the prediction/identification/removal operations are performed step-by-step to the 3 series, respectively. Each removal increases the density of the object-information, facilitating the next prediction. To comprehend the pixel distributions with their locations, a 2D histogram presentation is proposed. In the design of the brain-tumor detection system, it is used to highlight the left-right asymmetry of a brain structure. Since the asymmetry is caused by tumors and non-pathological elements, an adaptive histogram modulation method is proposed to enhance the former by attenuating the latter. The prediction/identification/removal operations transform a 3D brain image into a tumoral minimum bounding box, which is then transformed into a tumor mask using simple morphological operations. The test results, on 1251 samples, have confirmed the high quality of the prediction of the tumor data distributions and the tumor detection.
The second methodology is proposed to design CNN (convolutional neural network) systems handling a complex task of brain-tumor segmentation, i.e., classifying the pixels of a brain image into 4 classes of intra-tumoral regions and the background. The methodology is to decompose this complex task into simple subtasks and each of them is performed by a simply-configurated and independently-trained CNN. By doing so, one can optimize the use of computing power and minimize the gradient conflict in training. The 4-class classification is decomposed into 3 binary classifications. Each of them is further decomposed into 2: first locating the object region and then identifying the pixels inside the region, performed by 2 independent-and-simple modules. The overall system, requiring only 0.75M trainable parameters, has been trained/tested with BraTS datasets, and its processing quality is among the best reported recently
XFEM for the Homogenization of Constitutive Properties and its Stabilization for the Computation of Interface Tractions
Get PDFA computational homogenization method for assessing the local stress distributions and effective mechanical characteristics of composite materials is presented in this work. Using the (XFEM) modeling capabilities, the effects of bonding assumptions on interface tractions, local stress fields, and homogenized constitutive properties are examined. The possibility of standard XFEM-based models to generate spurious oscillatory tractions at the bond interfaces is shown, with interface stiffness affecting the oscillations' magnitude. A stabilized XFEM framework is suggested as a solution to this problem in order to lessen the oscillatory behavior at matrix-inclusion interfaces. Comparisons with results from other modeling methodologies that are accessible in the literature provide a thorough validation of the developed stabilizing strategy. The suggested methodology's resilience and adaptability are further illustrated by its application to a number of Representative Volume Element (RVE) cases. The impacts of mesh refinement, interface stiffness, and RVE boundary conditions on the precision and effectiveness of the suggested method are investigated parametrically. The findings show that the stabilized XFEM may produce dependable traction results across bond interfaces and emphasize the significance of interface stiffness in the precise prediction of stress distributions. Additionally, it is demonstrated that the suggested approach offers reliable and effective results for a variety of composite microstructural configurations. By providing a reliable tool for the homogenization and analysis of composites with imperfect interfaces under various mechanical loading situations, this work advances XFEM-based computational approaches
On Analyzing SSO Permissions Across Web and Android Platforms
Get PDFFederated Single Sign-On (SSO) is a widely used authentication method that delegates user login to Identity Providers (IdPs) such as Google and Facebook. While convenient, SSO raises privacy and security concerns, particularly, as we observed, when permissions vary across different platforms (web vs.\ mobile, even different versions of an app). Existing work on SSO logins completely lacks the exploration of such variances, and their privacy consequences, even though many users may use a service both via web and mobile platforms. This study examines such discrepancies at scale, alongside an analysis of dangerous permissions specifically requested on websites and Android apps. We developed a framework to automate SSO logins on both platforms, systematically measuring permission discrepancies. Our analysis, based on 661 and 318 successful logins using Google and Facebook SSO, respectively, across both the Android app and its corresponding website for the same service, reveals a 12.58\% discrepancy in Facebook SSO permissions and a 3.48\% discrepancy in Google SSO permissions between web and Android platforms. These findings, along with our analysis of top-5K Tranco websites, indicate that Android apps tend to request more intrusive permissions, underscoring the need for incremental authorization mechanisms to minimize unnecessary data exposure
Situated Embodiments: A Latourian Interpretation of Transcultural Psychiatry in Montreal
Get PDFThis thesis involves an ethnography of three field sites related to the scientific discipline of transcultural psychiatry — sites representative of its classroom, laboratory, and clinical activities. This psychiatric discipline, which incorporates conceptual and methodological elements of anthropology, holds that cultural variability and social positionality are important factors in illness and wellness. As such, the field disrupts both the materialist notions of the body dominant in contemporary biomedicine, and the cognitive explanatory models of conventional psychology. Despite this alternative approach to health and healthcare, transcultural psychiatrists in Montreal continue to position themselves as scientifically credible and medically authoritative.
My research project asks how the role and status of the body in medicine is affected in the context of such a field, which openly troubles received divisions between subjectivity and objectivity, nature and culture, social and material reality, and physiological or psychological stress. In particular, I consider the diagnostic attitudes and practices of transcultural psychiatry, which I articulate as an attempt to give name to culturally-situated forms of embodied disorder. By introducing ethnographic consultation into the diagnostic process, transcultural psychiatry opens up these scientific names to a new kind of participatory construction. This being the case, the diagnostic practice of the discipline consists of naming otherwise, or naming differently
Towards Efficient Device-State Integrity Verification in Smart Homes using Device-App Causality Relationships
Get PDFThe device-state in smart homes depends on both its physical channel (sensing and actuating in the environment) and its cyber-physical channel (interactions with apps and other devices). Ensuring device-state integrity is crucial for proper operation but can be compromised by security threats from devices, apps, and their interactions due to vulnerabilities and misconfigurations, posing risks to users. Existing works focus on either devices or apps, but none comprehensively address device-state integrity across device-app interactions. Furthermore, there exist several challenges in offering device-state integrity verification for smart homes. First, efficiently and comprehensively collecting data (that is an essential verification step) becomes more difficult as code instrumentation (used in several existing works) becomes impossible by changes in platform design and existing logging mechanisms (if any) generate humongous amount of data (including security unaware data). Second, efficiently and accurately verifying the device state integrity needs device-specific analysis to account for all interacting apps and devices. In this thesis, we tackle these challenges by presenting an efficient device-state integrity verification approach for smart homes. Specifically, our key ideas are to: model the interactions of all components in a smart home using causality relationships that affect a specific device, and verify device state based on this model. We implement our approach on SmartThings, build a new smart home dataset, and evaluate its effectiveness (e.g., 81.34% reduction in verification time and 56.49% reduction in response time)