Computer Engineering Commons^™

Improving Network Policy Enforcement Using Natural Language Processing And Programmable Networks, Pinyi Shi

Theses and Dissertations--Computer Science

Computer networks are becoming more complex and challenging to operate, manage, and protect. As a result, Network policies that define how network operators should manage the network are becoming more complex and nuanced. Unfortunately, network policies are often an undervalued part of network design, leaving network operators to guess at the intent of policies that are written and fill in the gaps where policies don’t exist. Organizations typically designate Policy Committees to write down the network policies in the policy documents using high-level natural languages. The policy documents describe both the acceptable and unacceptable uses of the network. Network operators …

Design, Development And Benchmarking Of Machine Learning Algorithms In Biomedical Applications, Qi Sun

Theses and Dissertations--Computer Science

Machine learning algorithms are becoming the most effective methods for knowledge discovery from high dimensional datasets. Machine learning seeks to construct predictive models through the analysis of large-scale heterogeneous data. While machine learning has been widely used in many domains including computer vision, natural language processing, product recommendation, its application in biomedical science for clinical diagnosis and treatment is only emerging. However, the wealthy amount of data in the biomedical domain offers not only challenges but also opportunities for machine learning. In this dissertation, we focus on three biomedical applications from vastly different domains to understand the opportunities and challenges …

Protocols And Architecture For Privacy-Preserving Authentication And Secure Message Dissemination In Vehicular Ad Hoc Networks, Shafika Showkat Moni

Theses and Dissertations--Computer Science

The rapid development in the automotive industry and wireless communication technologies have enhanced the popularity of Vehicular ad hoc networks (VANETs). Today, the automobile industry is developing sophisticated sensors that can provide a wide range of assistive features, including accident avoidance, automatic lane tracking, semi-autonomous driving, suggested lane changes, and more. VANETs can provide drivers a safer and more comfortable driving experience, as well as many other useful services by leveraging such technological advancements. Even though this networking technology enables smart and autonomous driving, it also introduces a plethora of attack vectors. However, the main issues to be sorted out …

An Automated Framework To Debug System-Level Concurrency Failures, Tarannum Shaila Zaman

Theses and Dissertations--Computer Science

The ever-increasing parallelism in computer systems has made software more prone to concurrency failures, causing problems during both pre- and post-development. Debugging concurrent programs is difficult because of the non-deterministic behavior and the specific sequences of interleaving in the execution flow. Debugging is a technique where programmers reproduce the bug, identify the root cause, and then take necessary steps to remove the bug from the system. The failure information may come from the bug reports of the testing phase or the production runs. In both cases, there should be steps taken to reproduce and localize the failure. However, reproducing and …

Ensemble Approach To The Semantic Segmentation Of Satellite Images, Brendan Kent

Dissertations

Automatic classification and segmentation of land use land cover(LULC) is extremely important for understanding the relationship between humans and nature. Human pressures on the environment have drastically accelerated in the last decades, risking biodiversity and ecosystem services. Remote sensing via satellite imagery is an excellent tool to study LULC. Research has shown that deep learning encoder-decoder architectures have achieved worthy results in the area of LULC, however the application of an ensemble approach has not been well quantified. Studies have shown it to be useful in the area of medical imaging. Ensembling by pooling together predictions to produce better predictions …

An Analysis On Network Flow-Based Iot Botnet Detection Using Weka, Cian Porteous

Dissertations

Botnets pose a significant and growing risk to modern networks. Detection of botnets remains an important area of open research in order to prevent the proliferation of botnets and to mitigate the damage that can be caused by botnets that have already been established. Botnet detection can be broadly categorised into two main categories: signature-based detection and anomaly-based detection. This paper sets out to measure the accuracy, false-positive rate, and false-negative rate of four algorithms that are available in Weka for anomaly-based detection of a dataset of HTTP and IRC botnet data. The algorithms that were selected to detect botnets …

Evaluating The Performance Of Vision Transformer Architecture For Deepfake Image Classification, Devesan Govindasamy

Dissertations

Deepfake classification has seen some impressive results lately, with the experimentation of various deep learning methodologies, researchers were able to design some state-of-the art techniques. This study attempts to use an existing technology “Transformers” in the field of Natural Language Processing (NLP) which has been a de-facto standard in text processing for the purposes of Computer Vision. Transformers use a mechanism called “self-attention”, which is different from CNN and LSTM. This study uses a novel technique that considers images as 16x16 words (Dosovitskiy et al., 2021) to train a deep neural network with “self-attention” blocks to detect deepfakes. It creates …

An Investigation Of The Relationship Between Subjective Mental Workload And Objective Indicators Of User Activity, Greg Byrne

Dissertations

Whilst the concept of physical workload is intuitively understood and readily applicable in system design, the same cannot be said of mental workload (MWL), despite its importance in our increasingly technological society. Despite its origin in the mid 20th century, the very concept of ”mental workload” is still a topic of debate in the literature, although it can be loosely defined as “the amount of mental work necessary for a person to complete a task” (Miller, 1956; Longo, 2014). Several methods have been utilized to measure of MWL, including physiological methods such as neuro-imagery, performance-based metrics, and subjective measures via …

Evaluating The Performance Impact Of Fine-Tuning Optimization Strategies On Pre-Trained Distilbert Models Towards Hate Speech Detection In Social Media, Aidan Mcgovern

Dissertations

Hate speech can be defined as forms of expression that incite hatred or encourage violence towards a person or group based on race, religion, gender, or sexual orientation. Hate speech has gravitated towards social media as its primary platform, and its propagation represents profound risks to both the mental well-being and physical safety of targeted groups. Countermeasures to moderate hate speech face challenges due to the volumes of data generated in social media, leading companies, and the research community to evaluate methods to automate its detection. The emergence of BERT and other pre-trained transformer-based models for transfer learning in the …

Examining The Effects Of Disabilities On Vr Usage And Accessibility Issues For Persons With Disabilities, Sean Williams

Dissertations

Virtual Reality (VR) is an emerging technology that’s popularity has been increasing at a yearly rate. Despite this, concerns about the accessibility of VR devices are ever-growing as many users struggle to use the technology, especially users with disabilities. This study analyses how different types of disabilities affect how often a user uses VR and any associated re-occurring difficulties that are related to specific types of disability. To do this, a previous survey regarding VR accessibility run by Disability Visibility Project and ILMxLAB is examined. In this survey, 79 participants who identify as having a disability answered questions related to …

Exploring 3d Data Reuse And Repurposing Through Procedural Modeling, Rachel Opitz, Heather Richards-Rissetto, Karin Dalziel, Jessica Dussault, Greg Tunink

Department of Anthropology: Faculty Publications

Most contemporary 3D data used in archaeological research and heritage management have been created through ‘reality capture,’ the recording of the physical features of extant archaeological objects, structures, and landscapes using technologies such as laser scanning and photogrammetry (Garstki 2020, ch.2; Magnani et al. 2020). A smaller quantity of data are generated by Computer Aided Design (CAD) and Building Information Modeling (BIM) projects, and even fewer data are generated through procedural modeling, the rapid prototyping of multi-component threedimensional (3D) models from a set of rules (Figure 8.1.). It is unsurprising therefore that in archaeology and heritage, efforts around digital 3D …

A Reynolds Number Based Sampling Technique For 3-D Vector Fields In Computational Fluid Dynamic Environments, Trent James Schweitzer

Graduate Student Theses, Dissertations, & Professional Papers

Effective visualization of unsteady, time-dependent vector fields in a virtual environment is not a trivial task. This is due to the fact that most visualization techniques require the user to have a prior understanding of how the vector field will behave to set the parameters used to create the visualization. In this thesis we will take air flow data from a computational fluid dynamic simulations to calculate the amount of turbulence (represented as Reynolds numbers) to identify regions of interest. We then calculate wind pathlines that will intersect with these points sampled from these regions. We address the issue of …

A Proportionality-Based Framework For Government Regulation Of Digital Tracing Apps In Times Of Emergency, Sharon Bassan

Dickinson Law Review (2017-Present)

Times of emergency present an inherent conflict between the public interest and the preservation of individual rights. Such times require granting emergency powers to the government on behalf of the public interest and relaxing safeguards against government actions that infringe rights. The lack of theoretical framework to assess governmental decisions in times of emergency leads to a polarized and politicized discourse about potential policies, and often, to public distrust and lack of compliance.

Such a discourse was evident regarding Digital Tracing Apps (“DTAs”), which are apps installed on cellular phones to alert users that they were exposed to people who …

Meltpondnet: A Swin Transformer U-Net For Detection Of Melt Ponds On Arctic Sea Ice, Ivan Sudakow, Vijayan K. Asari, Ruixu Liu, Denis Demchev

Electrical and Computer Engineering Faculty Publications

High-resolution aerial photographs of Arctic region are a great source for different sea ice feature recognition, which are crucial to validate, tune, and improve climate models. Melt ponds on the surface of melting Arctic sea ice are of particular interest as they are sensitive and valuable indicators and are proxy to the processes in the Arctic climate system. Manual analysis of this remote sensing data is extremely difficult and time-consuming due to the complex shapes and unpredictable boundaries of the melt ponds, and that leads to the necessity for automatizing the processes. In this study, we propose a robust and …

Circuit Optimization Techniques For Efficient Ex-Situ Training Of Robust Memristor Based Liquid State Machine, Alex Henderson, Christopher Yakopcic, Cory Merkel, Steven Harbour, Tarek M. Taha, Hananel Hazan

Electrical and Computer Engineering Faculty Publications

Spiking neural network hardware offers a high performance, power-efficient and robust platform for the processing of complex data. Many of these systems require supervised learning, which poses a challenge when using gradient-based algorithms due to the discontinuous properties of SNNs. Memristor based hardware can offer gains in portability, power reduction, and throughput efficiency when compared to pure CMOS. This paper proposes a memristor-based spiking liquid state machine (LSM). The inherent dynamics of the LSM permit the use of supervised learning without backpropagation for weight updates. To carry out the design space evaluation of the LSM for optimal hardware performance, several …

Development Of An Explainability Scale To Evaluate Explainable Artificial Intelligence (Xai) Methods, Stephen Mccarthy

Dissertations

Explainable Artificial Intelligence (XAI) is an area of research that develops methods and techniques to make the results of artificial intelligence understood by humans. In recent years, there has been an increased demand for XAI methods to be developed due to model architectures getting more complicated and government regulations requiring transparency in machine learning models. With this increased demand has come an increased need for instruments to evaluate XAI methods. However, there are few, if none, valid and reliable instruments that take into account human opinion and cover all aspects of explainability. Therefore, this study developed an objective, human-centred questionnaire …

Introduction To The Special Issue On Gala Conf 2021, Francesca De Rosa, Jannicke Baalsrud Hauge, Pierpaolo Dondio, Isa Marfizi-Schottman, Margarida Romero, Francesco Bellotti

Articles

This special issue of the International Journal of Serious Games is dedicated to the selected best papers of the 2021 edition of the GALA conference. This edition was organized by Francesca De Rosa and her team at NATO Centre for Maritime Research and Experimen-tation (CMRE), La Spezia, Italy. Because of the Covid-19 pandemic, it was held online, for the second year. The three papers published in this special issue were first selected for a content exten-sion, so to make them suitable as journal papers, then underwent the regular IJSG peer-review process, which, on the other hand, discarded other three selected …

Human Mental Workload: A Survey And A Novel Inclusive Definition, Luca Longo, Christopher D. Wickens, Gabriella Hancock, P.A. Hancock

Articles

Human mental workload is arguably the most invoked multidimensional construct in Human Factors and Ergonomics, getting momentum also in Neuroscience and Neuroergonomics. Uncertainties exist in its characterization, motivating the design and development of computational models, thus recently and actively receiving support from the discipline of Computer Science. However, its role in human performance prediction is assured. This work is aimed at providing a synthesis of the current state of the art in human mental workload assessment through considerations, definitions, measurement techniques as well as applications, Findings suggest that, despite an increasing number of associated research works, a single, reliable and …

The Association Between Ambient Uvb Dose And Anca‑Associated Vasculitis Relapse And Onset, Jennifer Scott, Enock Havyarimana, Albert Navarro-Gallinad, Arthur White, Jason Wyse, Jos Van Geffen, Michiel Van Weele, Antonia Buettner, Tamara Wanigasekera, Cathal Walsh, Louis Aslett, John Kelleher, Julie Power, James Ng, Declan O’Sullivan, Lucy Hederman, Neil Basu, Mark A. Little, Lina Zgaga

Articles

The aetiology of ANCA-associated vasculitis (AAV) and triggers of relapse are poorly understood. Vitamin D (vitD) is an important immunomodulator, potentially responsible for the observed latitudinal differences between granulomatous and non-granulomatous AAV phenotypes. A narrow ultraviolet B spectrum induces vitD synthesis (vitD-UVB) via the skin. We hypothesised that prolonged periods of low ambient UVB (and by extension vitD deficiency) are associated with the granulomatous form of the disease and an increased risk of AAV relapse.

Modeling Cognitive Load As A Self-Supervised Brain Rate With Electroencephalography And Deep Learning, Luca Longo

Articles

The principal reason for measuring mental workload is to quantify the cognitive cost of performing tasks to predict human performance. Unfortunately, a method for assessing mental workload that has general applicability does not exist yet. This is due to the abundance of intuitions and several operational definitions from various fields that disagree about the sources or workload, its attributes, the mechanisms to aggregate these into a general model and their impact on human performance. This research built upon these issues and presents a novel method for mental workload modelling from EEG data employing deep learning. This method is self-supervised, employing …

An Experimental Study Towards Underwater Propulsion System Using Structure Borne Traveling Waves, Shreyas Suhas Gadekar

Dissertations, Master's Theses and Master's Reports

The method of generating steady-state structure-borne traveling waves underwater in an infinite media creates abundant opportunities in the field of propulsive applications, and they are gaining attention from several researchers. This experimental study provides a framework for harnessing traveling waves in a 1D beam immersed under quiescent water using two force input methods and providing a motion to an object floating on the surface of the water.

In this study, underwater traveling waves are tailored using structural vibrations at five different frequencies in the range of 10Hz to 300Hz. The resulting fluid motion provides a propulsive thrust that moves a …

Multimodal Adversarial Learning, Uche Osahor

Graduate Theses, Dissertations, and Problem Reports (ETD)

Deep Convolutional Neural Networks (DCNN) have proven to be an exceptional tool for object recognition, generative modelling, and multi-modal learning in various computer vision applications. However, recent findings have shown that such state-of-the-art models can be easily deceived by inserting slight imperceptible perturbations to key pixels in the input. A good target detection systems can accurately identify targets by localizing their coordinates on the input image of interest. This is ideally achieved by labeling each pixel in an image as a background or a potential target pixel. However, prior research still confirms that such state of the art targets models …

High Energy And Power Density Peptidoglycan Musclesthrough Super-Viscous Nanoconfined Water, Haozhen Wang, Zhi-Lun Liu, Jianpei Lao, Sheng Zhang, Rinat Abzalimov, Tong Wang, Xi Chen

Advanced Science Research Center

Water-responsive (WR) materials that reversibly deform in response to humidity changes show great potential for developing muscle-like actuators for miniature and biomimetic robotics. Here, it is presented that Bacillus (B.) subtilis’ peptidoglycan (PG) exhibits WR actuation energy and power densities reaching 72.6 MJ m−3 and 9.1 MW m−3, respectively, orders of magnitude higher than those of frequently used actuators, such as piezoelectric actuators and dielectric elastomers. PG can deform as much as 27.2% within 110 ms, and its actuation pressure reaches ≈354.6 MPa. Surprisingly, PG exhibits an energy conversion efficiency of ≈66.8%, which can be attributed to its super-viscous nanoconfined …

Schaeffler F-250 Bev Conversion: Power Electronics, Leslie Sawyer

Williams Honors College, Honors Research Projects

This report shows the work that was done on the power electronics system in the industry sponsored Schaeffler demonstration truck project. The goal of the project was to convert a combustion Ford F250 into a battery electric vehicle, while using as much Schaeffler technology as possible. After the vehicle is converted it will be used as a demonstration vehicle to show potential investors. While this project is still in progress, the work that was done during this time was substantial to the development of this project. Progress that has been made on the power electronics system includes developments in selecting …

Novel Natural Language Processing Models For Medical Terms And Symptoms Detection In Twitter, Farahnaz Golrooy Motlagh

Browse all Theses and Dissertations

This dissertation focuses on disambiguation of language use on Twitter about drug use, consumption types of drugs, drug legalization, ontology-enhanced approaches, and prediction analysis of data-driven by developing novel NLP models. Three technical aims comprise this work: (a) leveraging pattern recognition techniques to improve the quality and quantity of crawled Twitter posts related to drug abuse; (b) using an expert-curated, domain-specific DsOn ontology model that improve knowledge extraction in the form of drug-to-symptom and drug-to-side effect relations; and (c) modeling the prediction of public perception of the drug’s legalization and the sentiment analysis of drug consumption on Twitter. We collected …

Garden Bot: Autonomous Home Garden Weed Removal Robot, Brendon Lovejoy, Robert Connolly, Isaac Lucas, Stevan Veselinov

Williams Honors College, Honors Research Projects

With frequent weeding being a tedious chore and an essential task for a successful garden, there is need for an automated method of handling this routine. Existing technologies utilize computer vision, GPS, multiple units and other tools to remove weeds from garden plots. However, these solutions are often complex and expensive, suited for large agricultural plots in contrast to small-scale home gardens. In addition, many of these technologies, along with manual tillers and cultivators suited for home use, are unable to perform weeding within rows of crops in a process known as intra-row weeding. The Garden Bot is an autonomous, …

A Solder-Defined Computer Architecture For Backdoor And Malware Resistance, Marc W. Abel

Browse all Theses and Dissertations

This research is about securing control of those devices we most depend on for integrity and confidentiality. An emerging concern is that complex integrated circuits may be subject to exploitable defects or backdoors, and measures for inspection and audit of these chips are neither supported nor scalable. One approach for providing a “supply chain firewall” may be to forgo such components, and instead to build central processing units (CPUs) and other complex logic from simple, generic parts. This work investigates the capability and speed ceiling when open-source hardware methodologies are fused with maker-scale assembly tools and visible-scale final inspection.

The …

Few-Shot Malware Detection Using A Novel Adversarial Reprogramming Model, Ekula Praveen Kumar

Browse all Theses and Dissertations

The increasing sophistication of malware has made detecting and defending against new strains a major challenge for cybersecurity. One promising approach to this problem is using machine learning techniques that extract representative features and train classification models to detect malware in an early stage. However, training such machine learning-based malware detection models represents a significant challenge that requires a large number of high-quality labeled data samples while it is very costly to obtain them in real-world scenarios. In other words, training machine learning models for malware detection requires the capability to learn from only a few labeled examples. To address …

Evaluating Similarity Of Cross-Architecture Basic Blocks, Elijah L. Meyer

Browse all Theses and Dissertations

Vulnerabilities in source code can be compiled for multiple processor architectures and make their way into several different devices. Security researchers frequently have no way to obtain this source code to analyze for vulnerabilities. Therefore, the ability to effectively analyze binary code is essential. Similarity detection is one facet of binary code analysis. Because source code can be compiled for different architectures, the need can arise for detecting code similarity across architectures. This need is especially apparent when analyzing firmware from embedded computing environments such as Internet of Things devices, where the processor architecture is dependent on the product and …

Locality Analysis Of Patched Php Vulnerabilities, Luke N. Holt

Browse all Theses and Dissertations

The size and complexity of modern software programs is constantly growing making it increasingly difficult to diligently find and diagnose security exploits. The ability to quickly and effectively release patches to prevent existing vulnerabilities significantly limits the exploitation of users and/or the company itself. Due to this it has become crucial to provide the capability of not only releasing a patched version, but also to do so quickly to mitigate the potential damage. In this thesis, we propose metrics for evaluating the locality between exploitable code and its corresponding sanitation API such that we can statistically determine the proximity of …