Computer Engineering Commons^™

A Personalized Chat Application For Career Profiling: A Case Study, Dhiraj Choithramani

Harrisburg University Dissertations and Theses

Artificial intelligence (AI) has rapidly transformed numerous fields over the past decade, significantly influencing industries such as software engineering and computer science. One of the most impactful developments in this area is the rise of AI-driven chat applications, which have evolved from simple, rule-based systems to sophisticated platforms capable of simulating human-like conversations. These chatbots are increasingly being utilized across various sectors, including customer service, healthcare, and education, to provide users with quick, personalized responses. This research paper presents a case study of a personalized chat application designed specifically for career profiling, leveraging advanced AI technologies to deliver contextually relevant …

A Step Towards Automated Ethical Analysis In Journalism: Measuring Llms’ Performance In Extracting Sourcing Information, Jingsen Wang

Computer Science and Engineering Master's Theses

This thesis explores the potential of Large Language Models (LLMs) in automating the extraction of sourcing information from news articles, a crucial step towards enhancing transparency and ethical analysis in journalism. We evaluate the performance of two state-of-the-art LLMs, GPT-4 and Claude 3, in identifying and categorizing various source types across four diverse news articles. The thesis employs a zero-shot learning approach with two different prompt designs, assessing the models’ ability to adapt to varying source structures and prompt instructions.

Our findings reveal that while LLMs show promise in extracting sourcing information, their performance varies significantly across different article types …

Activity Map Generation And Event-Based Sensor Processing With Spiking Autoencoders And Sparse Dictionary Learning, Jack Easton

Computer Science and Engineering Theses and Dissertations

This thesis explores the potential of Spiking Neural Networks (SNNs) in processing event sensor data and generating high-fidelity activity maps. Event sensors capture asynchronous binary events with high dynamic range, but traditional processing methods often fail to leverage their advantages fully. SNNs, with their asynchronous, event-driven nature, offer a promising alternative.

A Spiking Autoencoder (SAE) was employed in this thesis to handle the stochastic and sparse event data, integrating deep dictionary learning to enhance the feature space and improve activity map quality. The encoder, modeled after the VGG network, extracts features from event streams generated by speckle patterns, which are …

Evaluating The Role Of Data Enrichment Approaches Towards Rare Event Analysis In Manufacturing, Chathurangi Shyalika, Ruwan Wickramarachchi, Fadi El Kalach, Ramy Harik, Amit P. Sheth

Publications

Rare events are occurrences that take place with a significantly lower frequency than more common, regular events. These events can be categorized into distinct categories, from frequently rare to extremely rare, based on factors like the distribution of data and significant differences in rarity levels. In manufacturing domains, predicting such events is particularly important, as they lead to unplanned downtime, a shortening of equipment lifespans, and high energy consumption. Usually, the rarity of events is inversely correlated with the maturity of a manufacturing industry. Typically, the rarity of events affects the multivariate data generated within a manufacturing process to be …

Story Of Your Lazy Function’S Life: A Bidirectional Demand Semantics For Mechanized Cost Analysis Of Lazy Programs, Liyao Xia, Laura Israel, Maite Kramarz, Stephanie Weirich, Koen Claessen, Nicolas Coltharp, Yao Li

Computer Science Faculty Publications and Presentations

Lazy evaluation is a powerful tool that enables better compositionality and potentially better performance in functional programming, but it is challenging to analyze its computation cost. Existing works either require manually annotating sharing, or rely on separation logic to reason about heaps of mutable cells. In this paper, we propose a bidirectional demand semantics that allows for extrinsic reasoning about the computation cost of lazy programs without relying on special program logics. To show the effectiveness of our approach, we apply the demand semantics to a variety of case studies including insertion sort, selection sort, Okasaki's banker's queue, and the …

A Hierarchical Framework For Interpretable, Safe, And Specialised Deep Reinforcement Learning, Ammar Abbas

Doctoral

Safety-critical systems, which are crucial for human safety and the environment, are difficult to control and operate. Traditional controllers need precise models of these complex systems, which is hard to develop. \acrfull{drl} offers a potential solution by learning from interactions rather than detailed models, but it faces limitations such as non-transparent decision-making and an expensive, unsafe learning process. Additionally, a key challenge in DRL is ensuring effective decision-making in rare situations.

This thesis proposes a novel approach called the \acrfull{prop_frame} that enables safe and reliable control of critical systems. SRLA combines probabilistic modelling with reinforcement learning to create an interpretable …

Improving Expressive Capacity Of Deep Neural Networks, Clayton Harper

Computer Science and Engineering Theses and Dissertations

Deep learning has had remarkable success in a variety of fields. However, architectures often rely on hyperparameter searches and heuristics for improved model performance. Performing hyperparameter searches is an arduous task--often time-consuming and potentially expensive to run on accelerated hardware. As a result, practitioners often rely on heuristics which may lead to sub-optimal results. In the context of deep learning, hyperparameters are set by the user prior to the training process and remain fixed. Deep learning uses gradient descent to learn complex feature representations from data, limiting human intervention. While the weights of the architecture can learn directly through data …

Iot Technologies To Enable Location Based Services (Lbs) For Smart Tourism, Dharmik Chhatbar

University Honors Program Senior Projects

Location-based Services (LBS) are special types of services that are triggered to deliver certain content (e.g., text, image, video) to users with respect to the users’ exact or approximate location or proximity to certain points of interest. LBS can address the field of smart museums with a goal of enhancing the museum experience for visitors with dynamic and personalized content. In this UHP thesis, we explore a set of Internet of Things (IoT) technologies that can enable the work of LBS in the field of smart museums. The project envisions an immersive museum journey where visitors can seamlessly access interactive …

Vision-Based Autonomy Stacks For Farm Tractors And Intelligent Spraying Systems In Orchards, Shengli Xu

All Dissertations

Autonomous tractors equipped with intelligent sprayers have become a pivotal aspect of smart farming (SF), marking a transformative shift in traditional agricultural practices and holding the potential to revolutionize the farming industry. With 2,453,620 fruit-bearing acres in the United States as of 2022, there is a pressing need for the implementation of autonomous systems for farm tractors and intelligent spraying systems in orchards. These advancements can significantly reduce labor costs, address labor shortages, and minimize spray loss. Furthermore, to enhance profitability and productivity, it is essential to develop low-cost yet effective vision-based autonomy systems that can operate efficiently across various …

Convex Approach To Data-Driven Optimal Control With Safety Constraints Using Linear Transfer Operator, Joseph Raphel Moyalan

All Dissertations

This thesis is concerned with the data-driven solution to the optimal control problem with safety constraints for a class of control-affine nonlinear systems. Designing optimal control satisfying safety constraints is a problem of interest in various applications, including robotics, power systems, transportation networks, and manufacturing. This problem is known to be non-convex. One of this thesis's main contributions is providing a convex formulation to this non-convex problem. The second main contribution is providing a data-driven framework for solving the control problem with safety constraints. The linear operator theoretic framework involving Perron-Frobenius and Koopman operators provides the convex formulation and associated …

Patterning Synthesis Of Lead Halide Perovskites Toward Photonic Application, S. M. Nayeem Arefin

Department of Electrical and Computer Engineering: Dissertations, Theses, and Student Research

Lead halide perovskites (LHPs) are a fascinating class of photonic materials with the potential to revolutionize various optoelectronic applications. Their diverse crystal structures, ranging from 0D to 3D configurations, offer a unique combination of properties, including high tunability and ease of synthesis. However, their inherent instability and the difficulty of patterning them into sophisticated photonic structures using conventional methods present a significant hurdle to their widespread applications. This thesis addresses these challenges by proposing a novel synthesis method that combines soft lithography and self-assembly. By utilizing a patterned template with controlled wettability, precise manipulation of LHP crystal formation is achieved, …

Ultrashort Pulsed Laser Treatment Is Effective At Sterilizing Metal Surfaces For Planetary Protection, Kaleb Mcquillan

Department of Electrical and Computer Engineering: Dissertations, Theses, and Student Research

To prevent forward contamination from microbes aboard spacecraft intended for exploration of solar system bodies there is a need for effective sterilization methods. However, current techniques are both time-consuming and expensive. For example, dry heat sterilization requires removal from the assembly site and several days of treatment. Furthermore, some components such as optics and electronics are not compatible with current sterilization techniques. In this thesis, a novel femtosecond laser surface processing technique for the rapid sterilization of spacecraft hardware is reported. Femtosecond lasers produce extremely high photon fluxes (10²⁹ photons/s*cm², ~0.03 J/cm²) in extremely short …

Remodel-Fpga: Reconfigurable Memory-Centric Array Processor Architecture For Deep-Learning Acceleration On Fpga, Md Arafat Kabir

Graduate Theses and Dissertations

Deep-Learning has become a dominant computing paradigm across a broad range of application domains. Different architectures of Deep-Networks like CNN, MLP, and RNN have emerged as the prominent machine-learning approaches for today’s application domains. These architectures are heavily data-dependent, requiring frequent access to memory. As a result, these applications suffer the most from the memory bottleneck of the von Neumann architectures. There is an imminent need for memory-centric architectures for deep-learning and big-data analytic applications that are memory intensive. Modern Field Programmable Gate Arrays (FPGAs) are ideal programmable substrates for creating customized Processor in/near Memory (PIM) accelerators. Modern FPGAs contain …

Development Of Feature Extraction Models To Improve Image Analysis Applications In Cancer, Yu Shi

Dissertations and Doctoral Documents, University of Nebraska-Lincoln, 2023–

Cancer poses a significant global health challenge. With an estimated 20 million new cases diagnosed worldwide in 2022 and 9.7 million fatalities attributable to the disease, the economic burden of cancer is immense. It impacts healthcare systems and imposes substantial costs for its care on patients and their families. Despite advancements in early detection, prevention, and treatment that have reduced overall cancer mortality rates, the growing prevalence of cancer, particularly among younger individuals, remains a pressing issue.

Recent advancements in medical imaging technology have progressed significantly with the help of emerging computer vision and artificial intelligence (AI) technology. Despite these …

Development Of The Structure And Control System Of A Stewart Platform Robot For Human Balance Recovery Interventions, Rhobenn R. Alvarez Zambrano

Theses and Dissertations

In this thesis, the process to design a Stewart platform parallel robot for balance recovery with given assembly constraints and mobility requirements is described. A Model Based Design (MBD) approach in MATLAB was used as a tool to model and optimize the design of the platform through quick and repeatable workspace and movements simulations. An algorithm based on Inverse Kinematics was used to find the most adequate Stewart platform configuration which yields a workspace that fulfills the design goals best. Solidworks was used as a 3D CAD Modeling tool to elaborate machining blueprints while ensuring that each piece fits accurately …

Crowdstrike Cyber Incident Vs. Past Major Cyber Incidents: Analysis And Solutions, Priyant Banerjee

Himalayan Research Papers Archive

On July 19, 2024, a technical malfunction in CrowdStrike’s Falcon sensor software led to a global ITdisruption, affecting millions of devices across multiple sectors. This incident, although not a direct cyber-attack, caused significant operational upheavals reminiscent of major past cyber incidents. This paperexplores the CrowdStrike incident in detail, compares it with previous major cyber events, and proposescomprehensive solutions to mitigate such risks in the future.The faulty update from CrowdStrike resulted in widespread system crashes, notably the "Blue Screen ofDeath," paralyzing operations in critical sectors such as healthcare, finance, and transportation. The paperexamines the immediate and cascading effects of the incident, …

Future-Ready Digitalized Education: Unraveling The Dynamics Of Sustainable And Ethical Digital Transformation, Vaishnavi Rode

Electronic Theses, Projects, and Dissertations

Amid the brisk advancement of digital technologies, higher educational institutions and universities are finding themselves at a crucial turning point, with significant obstacles and new prospects in the realm of digital transformation. This culminating experience project delves deeply into the compounded terrain of digital transformation in higher education, emphasizing the need for sustainable practices in the face of rapidly evolving technical advancements. The research questions are: (Q1) What strategies can universities adopt to foster digital literacy among students and faculty while promoting sustainability values within their digital education programs and Why? (Q2) What ethical considerations, concerning data privacy and digital …

Task Management Application, Dhaval Chaturbhai Hirpara

Electronic Theses, Projects, and Dissertations

The Task Management Application is a web-based platform designed to facilitate efficient task and project management, similar to other Project Management Tools like Jira, Trello, ClickUp, Wrike, Zoho Projects, and Asana. The application features three distinct roles: Administrator, Project Manager, and Employee, each with specific functionalities and permissions to streamline workflow.

Administrator: This role encompasses comprehensive project oversight, including adding, viewing, and managing project managers, supervising ongoing projects, and viewing employee details.

Project Manager: Project Managers can manage employees, assign tasks, and oversee project progress effortlessly.

Employee: Employees have dedicated functionalities to view and manage tasks assigned …

Training Uav Teams With Multi-Agent Reinforcement Learning Towards Fully 3d Autonomous Wildfire Response, Bryce Hopkins

All Theses

As climate-exacerbated wildfires increasingly threaten landscapes and communities, there is an urgent and pressing need for sophisticated fire management technologies. Coordinated teams of Unmanned Aerial Vehicles (UAVs) present a promising solution for detection, assessment, and even incipient-stage suppression – especially when integrated into a multi-layered approach with other recent wildfire management technologies such as geostationary/polar-orbiting satellites and CCTV detection networks. However, there remains significant challenges in developing the necessary sensing, navigation, coordination, and communication subsystems that enable intelligent UAV teams. Further, federal regulations governing UAV deployment and autonomy pose constraints on real-world aerial testing, creating a disconnect between theoretical research …

Using Efficient Vision Transformers To Improve Perception Systems In Autonomous Off-Road Vehicles, Adam S. Pickeral

All Theses

The development of autonomous vehicles has become one of the greatest research endeavors in recent years. These vehicles rely on many complex systems working in tandem to make decisions. For practical use and safety reasons, these systems must not only be accurate, but also be quick to make decisions. In Autonomous Vehicle research, the environment perception system is one of the key com- ponents of development. The environment perception system allows the vehicle to understand its surroundings using cameras, light detection and ranging (LiDAR), and other sensor systems or modalities. Deep learning computer vision algorithms have shown to be the …

Adaptive Robot Collaboration Using Robotic Skin And Motion Similarity., Jordan Dowdy

Electronic Theses and Dissertations

An essential part of robotics research is human-robot collaboration, which enables the use of current and new robots in everyday life and the workforce. This research applies to both parts of human-robot collaboration: physical human-robot interaction (pHRI), as well as non-physical human-robot interaction. The physical interaction uses tactile sensors and a Neuroadaptive Controller (NAC) to allow for the guidance of a robotic arm and its end-effector. The non-physical interaction uses a novel motion similarity metric, the Cartesian Segment Online Dynamic Time-Warping (SODTW), to allow a robot to better adapt to the speed of the user performing the motion during imitation …

Development And Implementation Of A Gps-Agnostic Drone Localization System, Alex Peterson

Boise State University Theses and Dissertations

This research develops a GPS-denied state estimation system to localize and orient a drone for touch-based installations on power line towers and cables. As opposed to environments like underground tunnels or building interiors, our system effectively identifies and utilizes sparse landmarks such as towers, cables, and ground features. Our approach utilizes Simultaneous Localization and Mapping (SLAM) to create and reference three-dimensional maps in real time. Specifically, we employ Georgia Tech Smoothing and Mapping (GTSAM), proposed by Georgia Tech's BORG Lab, a factor graph-based data structure consisting of measurement factors and unknown pose variables that we are implementing for solving the …

Exploiting Physical Side-Channel Information For Offensive And Defensive Ends, Sisheng Liang

All Dissertations

Side-channel information consists of side effects of computation that range from microarchitectural to physical phenomena. Empirical studies have demonstrated the practical exploitability of these side effects in real-world systems for malicious attacks and effective defenses. In this dissertation, we discover, analyze, and exploit certain physical side-channel information for end-to-end attacks and defense across three studies.

In the first study, we demonstrate a new DNN model extraction attack named Clairvoyance that exploits certain far-field electromagnetic signals emitted from a GPU to steal DNN models several meters away from the victim machine, even with some physical obstacles in between. Using Clairvoyance, an …

Hardware-Oriented Protection And Acceleration For Machine Learning Application, Antian Wang

All Dissertations

The security of Machine Learning (ML) grows along with the development of high-performance models and expanding application scenarios. Numerous users are benefiting from the convenience brought by transformative ML applications. In the meantime, various attackers are trying to find vulnerabilities within ML deployment service models, thereby undermining the performance of ML and jeopardizing stakeholders’ interests. The dissertation focuses on the two aspects of secure ML applications: acceleration and protection. Homomorphic Encryption (HE) emerges as a widely recognized security primitive suitable for the cloud computing service model, where the computation can be performed over ciphertext without decryption. However, evaluations in the …

Efficient And Secure Data Transmissions In Emerging Heterogeneous Wireless Networks, Sihan Yu

All Dissertations

The widespread deployment of wireless devices facilitates the Internet of everything, greatly enhancing communication efficiency and improving people’s daily experiences. The interconnectivity of wireless devices relies on wireless communication technologies. With the development of various emerging communication technologies, wireless networks have become increasingly vast and complex, giving rise to numerous new challenges such as efficiency and security concerns.

In wireless networks, different devices may utilize different communication protocols, resulting in heterogeneous wireless networks. Communication among heterogeneous devices is challenging, often leading to conflicts and low communication efficiency when utilizing limited communication resources (e.g., spectrum resources). Moreover, the coexistence of heterogeneous …

Image Processing Techniques For Water Droplet Penetration Time And Contact Angle Estimation, Sai Balaji Jai Kumar

UNLV Theses, Dissertations, Professional Papers, and Capstones

Water droplet behavior on soil surfaces plays a critical role in numerous environmental processes, including soil erosion, hydrological dynamics, and ecosystem health. Accurate characterization of soil water repellency, quantified by parameters such as water droplet penetration time (WDPT) and contact angles (WDCA), is essential for informed decision-making in agricultural management, forestry practices, and land-use planning. Despite the significance of these parameters, challenges exist in reliably estimating them due to the complex and dynamic nature of soil-water interactions. This thesis address challenges in estimating WDPT and WDCA, by leveraging state-of-the-art image processing techniques and machine learning algorithms. The research focuses on …

Automated Measurement Of The Water Drop Penetration Time For The Analysis Of Soil Water Repellency, Danxu Wang

UNLV Theses, Dissertations, Professional Papers, and Capstones

In this work, we develop an innovative system for the automated measurement of Water Drop Penetration Time (WDPT) - a parameter that is conventionally used for evaluating soil water repellency (SWR). Increased SWR can be a reason for plant stress and poor crop yields, create a risk of potential water runoff and floods and thus can pose risks to life and property loss. Timely evaluation of soil conditions can save resources and win time for responding to environmental disasters. Manual measurements of WDPT are labor-intensive, subjective, tend to produce variability of outcomes, and also not always available in remote or …

A Real-Time Iot-Based Data Acquisition And Monitoring System For Photovoltaic Applications, Adam Barbosa, Hamza Mubarak, Fazel Mohammadi, Mohammad J. Sanjari, Mehrdad Saif

Electrical & Computer Engineering and Computer Science Faculty Publications

The transition to low-carbon energy systems, driven by climate change and fossil fuel scarcity, highlights technologies, such as Photovoltaic (PV) technology, for sustainable energy generation. This paper focuses on enhancing the efficiency of PV monitoring systems by leveraging Internet of Things (IoT) technology for accurate and real-time monitoring of essential parameters, such as voltage, current, and output power. Significant gaps in cost-effective and reliable IoT integration for PV monitoring are addressed, with an emphasis on predictive modeling. In this regard, a low-cost real-time IoT-based data acquisition and monitoring system for PV systems, as a proof of concept for future endeavors …

Real-Time Gun Detection In Video Streams Using Yolo V8, Harish Kumar Reddy Kunchala

Electronic Theses, Projects, and Dissertations

In this research, we advance the domain of public safety by developing a machine learning model that utilizes the YOLO v8 architecture for real-time detection of firearms in video streams. A diverse and extensive dataset, capturing a range of firearms in varying lighting and backgrounds, was meticulously assembled and preprocessed to enhance the model's adaptability to real-world scenarios. Leveraging the YOLO v8 framework, known for its real-time object detection accuracy, the model was fine-tuned to accurately identify firearms across different shapes and orientations.

The training phase capitalized on GPU computing and transfer learning to expedite the learning process while preserving …

Society Management App, Ruchit Rakholiya

Electronic Theses, Projects, and Dissertations

A comprehensive solution as native mobile application which is feasible economical and fast, which will establish the authenticity and reliability for society management overcoming the drawbacks of current system. In today's fast-paced technological ecosystem, the capacity to readily store and access information is becoming increasingly important. Residential societies, where individuals live together and manage collective resources, often require a large number of documents, registrations, vehicle parking records, and other forms of paperwork. The complexity and volume of these documents can lead to inefficiencies and frustrations among residents and management alike.