Computer Engineering Commons^™

Phishing Detection Using Continual Learning And Large Language Models, Gopi Prajeev Battula

Master's Projects

Adaptive phishing detection remains crucial as the nature of cyber-attacks changes over time, which renders static models obsolete. This project extends phishing detection through the implementation of continual learning approaches, namely Elastic Weight Consolidation (EWC) and Learning Without Forgetting (LWF) with RoBERTa, a Large Language Model (LLM) and compares the results of these approaches against GPT-4o-mini, another LLM. Our approach begins with fine-tuning RoBERTa on multiple phishing datasets to establish an effective baseline. EWC is then implemented to preserve vital model parameters based on their importance measured by the Fisher Information Matrix, while LWF uses knowledge distillation to retain prior …

On The Adversarial Robustness Of Quantized Neural Networks Against Common Adversaries In Time-Series Forecasting, Maanak Arora

Master's Projects

Real-world edge applications now use modern machine learning models which require both resource efficiency and robustness against adversarial threats. Deep neural networks which include time series forecasting models still face risks from adversarial perturbations while quantization techniques used for memory and compute efficiency create unpredictable robustness challenges. This project investigates the adversarial resistance of Long Short-Term Memory (LSTM) models after applying post-training quantization at three different precision levels: 16-bit floating point (FP16), 8-bit integer (INT8) and custom 4-bit quantization. The Jena Climate dataset serves as our main benchmark for training a fullprecision LSTM model followed by multiple quantization strategies which …

Physiotrack: A Gamified Physiotherapy System, Pranavi Chaturvedula

Master's Projects

Traditional physiotherapy methods tend to be non-interactive and provide little to no personalized instruction, even though physiotherapy is critical to stroke recovery. This thesis explores a fully adaptive, sensor-based, feedback architecture intended for stroke patients which remotely supervises movement and personalizes exercises enabled by multimodal sensors. The system uses filtering and windowed segmentation of accelerometer and skeletal data to compute features like jerk, speed, and joint movement angular range. A game engine applies accelerometer and skeletal features together with optimized, lightweight ML models to drive adaptive feedback, scoring, and difficulty adjustment. The architecture supports responsive continuous sensor streaming within the …

Large Language Model Powered Etl Pipeline Development A Project Report, Aditya Chandras

Master's Projects

This projects discusses the use of Large language model to simplify and automate Extract, Transform, Loading (ETL) development. Data engineering tasks oUen need technical experDse, thereby challenging non-experts and even industry professionals for such Dme intensive operaDons. The project tackles these challenges by employing quanDzaDon of a base Lllama-2-7b-Chat model using QLora technique. The lowered precision of the base model allow further execuDon on a limited hardware resource. The project further performs Supervised fine-tuning trainer (SFT) to perform fine-tuning, specializing the model for generaDng script for ETL tasks. The fine-tuned model is evaluated through a series of micro and comprehensive …

Suicidal Ideation Detection On Reddit Using Llm-Annotated Data And Graph Neural Networks, Ikbal Singh Gurdev Singh Dhanjal

Master's Projects

Suicide is the fourth leading cause of death among people aged 15-29. More than 720, 000 people commit suicide every year. During the COVID-19 pandemic, we saw an increase in people seeking out mental health support on anonymous forums like Reddit. These anonymous forums allow people to express their suicidal ideation without judgment and give them a support structure that not everyone has. The aim of this project is to detect suicidal ideation using Reddit. In this work, we propose SIRGEL (Suicidal Ideation on Reddit using Graph Embeddings and LLMs), a dual-pipeline approach that combines large language model (LLM)- based …

Secured Data Storage Management With Deduplication In Cloud Computing And Local Gpt Integration, Pavan Myana

Master's Projects

Exponential growth in cloud computing has brought enormous changes in data storage and processing, but also raised several questions on the security, privacy, and efficient storage of data. This report provides a dual-focused approach toward solving these challenges. First, we try to build an application securely and efficiently using data deduplication and Proxy Re-Encryption for optimization of storage and enabling secure data sharing. Deduplication ensures that redundant data is removed before encryption for maximum efficiency in storage, while PRE enables the safe sharing of encrypted data by re-encrypting the keys for specified recipients without the leakage of sensitive information. We …

Performance Comparison Of Machine Learning Across Metal, Cuda, And Neuromorphic Frameworks, Ryan Saini

Master's Projects

Machine learning’s computational demands necessitate optimal performance and utilization. This research compares Apple Silicon M3 Pro with MPS, NVIDIA RTX 3070 GPU with CUDA, and neuromorphic computing for machine learning methods. We provide a cross-platform and cross-architecture performance analysis of machine learning methods to identify optimal configurations for training and inference scenarios. On traditional neural networks, Apple Silicon with MPS delivers superior energy efficiency at the cost of longer processing times for training and inference. NVIDIA with CUDA offers faster computation in training and inference at higher energy costs. Convolutional spiking neural networks perform competitively on event-based data, particularly on …

Malware Opcode Embedding And Quality Assessment Of Generative Sample Embeddings, Atishay Jain

Master's Projects

Malware is software used to damage and disrupt computer systems with the intent to cause damage to the victim. Malware detection and classification into malware families is a crucial problem for cybersecurity researchers. One of the major bottlenecks in improving these systems is the shortage of good quality labeled malware data, especially for malware families with scarce samples. Researchers have utilized generative models to generate malware data to address this issue. Malware embeddings encode patterns within a malware file, which can be used to detect and classify malware. Recently, encouraging results have been obtained in generating malware embeddings using generative …

Gen Ai For Malicious Network Data, Aneesh Maturu

Master's Projects

Though botnet attacks are on the rise, they also have become sophisticated and difficult to detect. Such a rising threat demands more and more sophisticated cybersecurity that leverages machine learning technology. Nevertheless, one of the biggest bottlenecks remains the unavailability of large and well-balanced datasets, particularly for malicious traffic, which hampers the efficacy of detection models. In an attempt to address this issue, our research utilizes Generative Adversarial Networks (GANs) to produce synthetic samples of botnet traffic from the CTU-13 dataset. While the majority of generative models have been targeting image data, we use GANs for a new application: generating …

Human Vs. Llm: Interpreting Emotion From Neutral Faces In Contextually Charged Backgrounds, Sai Mounika Peteti

Master's Projects

In this study, we investigate the intersection of cultural context, visual perception, and implicit bias in interpreting neutral human expressions. Specifically, we explore how emotionally charged backgrounds can shape viewers interpretations of neutral facial expressions. Using an experimental setup, participants are shown neutral human portraits paired with varying background types ranging from emotionally neutral or pleasant to evocative scenes. Participants are tasked with selecting which background best matches the emotional state of the person depicted in the portraits. Eye-tracking data is collected to analyze visual attention patterns and cognitive processing. In parallel, we also evaluate how large language models (LLMs) …

Vision-Language Models For Future Image Caption Prediction: Methods, Applications, And Adversarial Vulnerabilities, Md Ishak

Wayne State University Theses

Image captioning has traditionally focused on generating descriptions for individual static images. However, predicting future events from visual information is a fundamental challenge in this domain. While existing methods primarily describe current visual content, the ability to anticipate and generate captions for future events remains largely unexplored. We propose a novel approach for future caption prediction by leveraging the capabilities of Vision Transformer (ViT), Generative Pre-trained Transformer 2 (GPT-2), and Text-to-Text Transfer Transformer Model (T5) architectures.

Our method includes two complementary strategies: a two-stage pipeline where ViT-GPT2 generates captions for current images and T5 analyzes these captions to predict future …

Indigenous Technology Futurisms: Reclaiming Mackinac Island In Virtual Reality, Madeline A. Gupta

Library Map Prize

In Indigenous Technology Futurisms: Reclaiming Mackinac Island in Virtual Reality, Madeline A. Gupta introduces a place-based digital storytelling project that uses interactive mapping as the foundation for cultural reconnection and community wellness. Centered on Mackinac Island—ancestral land of the Anishinaabe people—the project features a custom-designed map by an Anishinaabe artist, allowing users to navigate a VR website by selecting specific locations across northern Michigan. Each map point opens into a spatial video experience paired with traditional audio content, including oral histories, poetry, and songs.

The project reframes digital cartography through an Indigenous lens, challenging Western colonial mapping practices that …

C 3 An: Custom, Compact And Composite Ai Systems - A Neurosymbolic Approach: 4Th-Generation Evolution Of Intelligent Systems, Amit P. Sheth, Kaushik Roy, Revathy Venkataramanan, Venkatesan Nadimuthu

Publications

Artificial Intelligence (AI) systems continue to evolve rapidly. From the architecture perspective, it is evolving from large, monolithic models trained on massive internet data to complex, multi-component “compound” systems and “agentic” frameworks capable of semi-autonomous decision-making. These systems show immense promise yet face numerous challenges in reliability, consistency, transparency, and alignment with user goals. In this article, we propose Custom, Compact and Composite AI with Neurosymbolic (C3AN) approach, a framework that paves way to 4th-generation of AI that integrates data, knowledge, and human expertise to build robust, intelligent and trustworthy AI systems defined by 14 foundation elements.

Custom emphasizes …

Hypnosis And Mindfulness Audio Recordings For Reducing Fatigue In Individuals With Multiple Sclerosis: A Randomized Controlled Study, Mark P. Jensen, Susan Robles, Michael G. Nash, Susanne May, Dwan M. Ehde, Melissa A. Day, Owen Gottlieb, Laurence I. Sugarman, Kevin N. Alschuler

Articles

Background

Fatigue is a common problem in individuals with multiple sclerosis (MS).

Objective

The objective was to evaluate the effects on fatigue of having 4 weeks of

access to audio recordings of therapeutic hypnosis (HYP) and mindfulness meditation

(MM) practices.

Methods

A total of 333 individuals with MS and fatigue were randomly assigned to

one of the three treatment conditions for 28 weeks: (1) access to therapeutic HYP audio

recordings, (2) access to MM audio recordings, or (3) no access to recordings

(treatment as usual or TAU). Fatigue impact (primary outcome) and other outcomes

were assessed at 4, 16, and …

Tracking Control Of A String Actuated Soft Trunk Robot Using State-Space Modeling, Jacob Trivisonno

Open Access Master's Theses

Soft robots, primarily composed of compliant materials, exhibit highly complex and nonlinear dynamics, making precise control a significant challenge. Traditional control methods often struggle with these complexities, requiring more advanced and data-driven methods. In our previous work, “Automatic Control of a Soft Trunk Robot Actuated by Strings” [1], a proportional controller was developed to drive the steady-state tracking error to zero. While this method was effective, it required time-consuming manual tuning and was not easily adaptable to physical modifications of the robot.

To address these limitations, this research implements gain-scheduled feedback control with state-space models calculated through a multiple linear …

Earthquake Wrangler: Leveraging Ios Technology For Earthquake Detection And Early Warning Application To Enhance Public Safety., Luis F. Salome

Theses and Dissertations--Civil Engineering

Earthquakes are devastating natural phenomena and generate secondary hazards such as tsunamis, landslides and fires. Their catastrophic impacts span both developed nations including the United States, Japan, Turkey, and Italy and developing countries such as El Salvador, Haiti, Nepal and the Philippines, where disparities in early warning infrastructure remain important. Seismic events start with stress waves generated by tectonic plate motion, with body waves (P- waves and S-waves) and surface Rayleigh and love waves that carry energy through the earth. While some regions have adopted advanced early warning systems based on seismic hazard models and strong ground motion analysis, others …

Augmenting Machine Learning Technique Through Natural Language, Tasmia Tasrin

Theses and Dissertations--Computer Science

While artificial intelligence (AI) and machine learning (ML) have proven effective at addressing many of the challenges that we face in our everyday lives, there are many situations in which these methods struggle. Examples include environments where AI or ML systems must perform complex behaviors or those where rewards are difficult to calculate. To address this limitation, interactive machine learning (IML) techniques have been introduced, which incorporate machine-understandable human feedback into traditional ML approaches. This feedback is often given as a discrete, positive or negative numeric value. This feedback is typically provided as often as possible to convey a dense …

Integrated Compliant Structure For A Hand Exoskeleton, Tristan R. Koopman

Honors Undergraduate Theses

This thesis presents the design and prototyping of a wearable hand exoskeleton that integrates a flexible structural framework to assist with hand movement while maintaining comfort and anatomical conformity. The goal was to create a device that supports tendon-driven actuation through a compliant structure, combining elements of rigidity and flexibility to match the natural geometry and motion of the human hand. Traditional hand exoskeletons often trade off motion for structure or vice versa. This project aims to bridge that gap with a hybrid compliant design that balances flexibility and support. The design process followed an iterative approach involving rapid prototyping …

Facial Chick Sexing: An Automated Chick Sexing System From Chick Facial Image, Marta Veganzones Rodriguez, Thinh Phan, Arthur F.A. Fernandes, Vivian Breen, Jesus Arango, Michael T. Kidd, Ngan Le

Computer Science and Computer Engineering Faculty Publications and Presentations

Chick sexing, the process of determining the gender of day-old chicks, is a critical task in the poultry industry due to the distinct roles that each gender plays in production. While effective traditional methods achieve high accuracy, color, and wing feather sexing is exclusive to specific breeds, and vent sexing is invasive and requires trained experts. To address these challenges, we propose a novel approach inspired by facial gender classification techniques in humans: facial chick sexing. This new method does not require expert knowledge and aims to reduce training time while enhancing animal welfare by minimizing chick manipulation. We develop …

Focused Feature Extraction For Driver Drowsiness Detection Using An Enhanced Attention-Based Resnet Model, Nada Ayman Atia, Rawan Sameh

The Undergraduate Research Journal

In the context of increasing road safety concerns, particularly in Egypt, this paper addresses the critical issue of driver drowsiness, a significant contributor to road accidents worldwide. With alarming statistics from the World Health Organization citing human error, chiefly drowsiness, as the cause for a majority of road accidents in Egypt, there is a compelling need for an effective drowsiness detection system. This research introduces a novel, vision-based driver drowsiness detection system leveraging a multi-dimensional approach with a Residual Neural Network (ResNet) architecture and attention layers. This system aims to accurately identify drowsiness by analyzing key facial features. The paper …