Engineering Commons^™

A Hands-On Lab On Agentic Scientific Discovery For Time Series With The Model Context Protocol, Dhaval Patel, Miaomiao Feng, Chathurangi Shyalika, Shuxin Lin, Hasnat Md. Abdullah

Publications

Agentic AI offers new opportunities to accelerate scientific discovery for time series by enabling autonomous systems to inspect data, formulate hypotheses, compose experiments, and refine models based on empirical evidence. However, building agents that make sound modeling decisions requires more than connecting a language model to forecasting tools. Key challenges include exposing large model catalogues without overwhelming the agent, grounding parameter choices in data characteristics, supporting iterative experimentation, and evaluating the quality of the reasoning behind each decision.

This hands-on lab introduces participants to agentic scientific discovery for time series using the Model Context Protocol (MCP). Through five interactive sprints, …

Domain Knowledge-Informed Machine Learning Model To Study Weather-Induced Flight Delays, Gurvir Bawa

Doctoral Dissertations and Master's Theses

Weather-induced Traffic Management Initiatives (TMIs) occur rarely relative to normal air traffic operations. This scarcity produces severe class imbalance that limits predictive models of flight delay. To address it, this dissertation leverages a physics- and domain- knowledge informed Variational Autoencoder (VAE), conditioned on airport identity, that generates physically plausible synthetic weather events for underrepresented TMI categories. The fidelity of the model is assessed with a calibrated discrepancy measure disclosed along- side the results rather than described in free-form terms. A multitask Graph Neural Network (GNN) is then trained on the augmented dataset to jointly classify airport-level TMI status and predict …

Aidws3: Stubborn And Hallucinating Agents, Inal Mashukov

Paul English Applied Artificial Intelligence (AI) Institute Publications

This workshop, part of the AI Demo Day Workshop Series hosted by the Paul English Applied AI Institute (PEAAII) at UMass Boston, explores the bias-variance tradeoff and how it affects the performance and reliability of AI models and agents. Led by Inal Mashukov, PhD Researcher at UMass Boston, the session covers concepts such as underfitting, overfitting, and regularization techniques to improve model performance. Through practical examples and hands-on exercises using PyTorch, participants explore why AI agents may hallucinate, repeat mistakes, or struggle to adapt, and learn approaches to building more reliable AI systems.

Design And Fabrication Of 3d Bioprinted Scaffold: Experimental And Machine Learning Methods, Mohan K. Dey

LSU Doctoral Dissertations

The development of reliable hydrogel-based scaffolds for extrusion bioprinting remains limited by the poor structural fidelity of low-viscosity bioinks and the lack of robust, high-throughput quality evaluation methods. This dissertation can resolve such challenges by developing the combination of optimized hydrogel formulations, cryogenic-assisted bioprinting, and artificial intelligence (AI)-based scaffold evaluation to the use of tissue engineering and preclinical cancer modelling applications. To assess the rheological behavior, printability, mechanical properties, and biocompatibility of the alginate – gelatin (Alg–Gel) system, a novel system of hydrogel was developed and characterized using Alg–Gel hydrogel system. The 7% alginate, 8% gelatin mixture was found to …

Validated Near-Infrared Spectroscopy And Chemometric Modelling For Rapid Quantification Of Essential Oil Yield And Α/Β-Santalol In Santalum Album L., Muhammad Hassnain, Muhammad Rizwan Azhar

Research outputs 2022 to 2026

The sandalwood industry remains constrained by destructive, time-intensive assays for essential oil (EO) yield, composition, moisture content, and wood fraction, which limit real-time decision-making. We report a unified near-infrared spectroscopy-artificial intelligence (NIRS-AI) platform for non-destructive analytics across the Santalum album L. value chain. Reflectance and transmittance spectra from solid matrices (disks, logs, chips, and powders), oils, ethanol extracts, and CID-derived emulsions were acquired using benchtop (400–2500 nm) and portable (900–1700 nm) spectrometers and calibrated against hydrodistillation, gas chromatography, extraction, and moisture assays. Advanced chemometric modelling using AI-based machine learning techniques, including regularised regression, ensemble learning, boosting algorithms, and neural networks, …

Evaluating Machine Learning Models On Classification Of Novel Cyber Attacks In The Healthcare Domain, Promise Ehimen

Dissertations, Theses, and Projects

The increasing adoption of the Internet of Medical Things (IoMT) has improved healthcare delivery through connected medical devices while simultaneously expanding the cybersecurity risks facing healthcare organizations. Although machine learning based intrusion detection systems have demonstrated high detection accuracy, their ability to respond reliably to previously unseen cyberattacks remains uncertain. This study investigated how a Neural Network model and a Logistic Regression model classified novel cyberattacks within the IoMT environment. The Neural Network and Logistic Regression models were both trained and tested using a subset of the CICIoMT2024 benchmark dataset. The Neural Network achieved 99.82% test accuracy and a 0.94 …

Explainable Machine Learning For Biomedical Diagnostics: Optical Imaging And Eeg Signal Analysis, Fozia Rajbdad

LSU Doctoral Dissertations

The growing convenience of complex biomedical data begins new roads for better disease detection and functional identification via artificial intelligence (AI). Nevertheless, conventional analysis methods often rely on basic metrics that drop sensitive biotic differences, and various AI systems are difficult to infer, limiting their clinical reliability and practical use. There is a growing need for explainable, physiologically relevant computational models that can extract key biomarkers from diverse biomedical data sources. This dissertation addresses this problem by obtaining explainable machine learning and deep learning procedures for studying biomedical signals and optical imaging data.

This dissertation is divided into two parts; …

Feasibility Of Using Untreated And Treated Rice Straw Fibers As A Modifier For Hot Mix Asphalt, Aditya Chowdhury

Civil Engineering Theses

Flexible pavements are subject to progressive deterioration driven by repeated traffic loading, environmental exposure, and material degradation, which causes distress. This study presents an integrated research framework that combines laboratory performance evaluation of rice straw fiber (RSF)-modified hot mix asphalt with machine learning-based pavement distress prediction to advance sustainable pavements. Three plant-produced HMA mixtures were modified with untreated and alkali-treated RSF at dosage levels of 0.2%, 0.4%, and 0.6% by total mixture weight. FTIR spectroscopic analysis confirmed that alkali treatment induced significant chemical modifications to the fiber surface, increasing hydroxyl group availability and enhancing fiber–binder adhesion. Alkali-treated RSF consistently outperformed …

Comprehensive Study On Explainable Artificial Intelligence For Enhanced Decision-Making, Eman Taher, Wessam H. El-Behaidy, Doaa S. Elzanfaly

Computer Science

Artificial Intelligence (AI) models are often criticized for their black-box nature, particularly in high-stakes domains such as finance, healthcare, and business decision-making, where transparency, accountability, and trust are essential. As machine learning advances with more complex architectures, including deep neural networks and ensemble models, the need for explainability becomes increasingly critical. This study focuses on Explainable Artificial Intelligence (XAI) as a key approach to enhance interpretability in classification, regression, and clustering tasks that serve as the foundation of data-driven analytical systems. XAI methods such as SHAP (SHapley Additive exPlanations), LIME (Local Interpretable Model-agnostic Explanations), and Integrated Gradients provide a means …

Automated Optimization Of Membrane Bioreactor Performance Using Machine Learning, Janessa Cayabyab, Isabella Lopez, Oluwatomisin Opaleye, Jessica Stone

Bioengineering Senior Theses

Membrane bioreactors (MBRs) are increasingly employed in biomanufacturing due to their ability to produce high-quality effluent and support complex biological processes. However, membrane fouling remains a critical operational challenge, driving increased energy consumption, frequent cleaning cycles, elevated transmembrane pressure (TMP), and accelerated membrane degradation. Conventional operator-driven cleaning strategies are reactive by nature and insufficient for managing the dynamic, nonlinear fouling behavior characteristic of full-scale MBR systems.

This study presents a data-driven machine learning pipeline for predicting TMP and fouling behavior in an industry-scale MBR system, with the goal of optimizing cleaning schedules and extending membrane lifespan. A Random Forest regression …

Evaluating Soil Health And Crop Yield In Louisiana Agricultural Systems: Impacts Of Best Management Practices And Prediction Models, Hector J. Mendoza Lagos

LSU Doctoral Dissertations

The adoption of conservation management practices is critical for improving soil health, enhancing nutrient use efficiency, and sustaining crop productivity in row crop systems in Louisiana. This study evaluated the role of conservation agronomic practices, soil biochemical indicators, and machine learning predictive models to improve soil nutrient dynamics, soil health indicators, microbial communities (MC), and crop productivity on a corn (Zea mays L.) research plot scale and in a commercial forty-hectare cotton (Gassypium hirsutum L.)-corn-soybean (Glycine max L.) rotation system in northeast Louisiana. The objectives of the study were to evaluate soil nutrient dynamics and MCs under …

Machine Learning And Formal Methods In Quantum Chemistry: Theory And Application, Ishna Satyarth

Computer Science and Engineering Theses and Dissertations

In recent years, the progress in inter-disciplinary application of machine learning and artificial intelligence (ML/AI) have truly transformed various fields, from weather forecasting and drug development to medical diagnostics, energy, and sustainability. Computational chemistry uses computational tools to model, predict, analyze, and explain chemical phenomena, while the Quantum chemistry specifically uses techniques based on quantum mechanics (as opposed to classical mechanics or empirical models). Quantum chemistry or Computational chemistry has also observed a momentum in application of ML techniques over the past decade significantly accelerating results and providing valuable insights into vast datasets, often surpassing traditional methods.

This dissertation explores …

Machine Learning Assisted Development Of Al0.2cufemnni High Entropy Alloy Through Selective Laser Melting, Hareharen K Mr

Theses and Dissertations

High Entropy Alloys (HEAs) are an emerging class of advanced materials that have gained significant attention due to their exceptional mechanical strength, thermal stability, and structural performance. Unlike conventional alloys based on a single principal element, HEAs are composed of multiple elements in a near-equiatomic ratio. Despite these advantages, designing HEAs with tailored properties is difficult because of the enormous number of possible combinations and the limitations of traditional trial-and-error methods. To overcome these challenges, this study presents a machine learning (ML) based approach to accelerate the design and development of an HEA.

In this work, a newly designed composition, …

Analysis Of Dynamic Difficulty Scaling Ai In Video Games, Tyrese W. Walker

Honors Program: Senior Projects (Public)

Video games are a popular form of interactive media and, since their inception, have become one of the largest forms of media consumed. As such, they have evolved greatly from their humble beginnings into much more complex experiences, and adjusting the difficulty level to suit the needs of the player has become common practice. While there are simple ways to do so, the best games often feature a dynamic difficulty-scaling system that adapts to the player. Classics like Resident Evil and Left 4 Dead are excellent examples of innovative dynamic scaling design. By analyzing the strongest elements of these works, …

Ais26s: Ai In Biomedicine, Shiqian Shen

Paul English Applied Artificial Intelligence (AI) Institute Publications

This presentation explores the role of artificial intelligence in advancing biomedical research and clinical practice from the perspective of a physician-scientist. Dr. Shiqian Shen discusses current challenges in neuroscience and medicine, including limitations in human observation, data analysis, and decision-making across complex biological systems. The talk highlights how AI-driven approaches such as computer vision, neural signal processing, and large-scale data modeling can improve the understanding of disease mechanisms, enhance experimental workflows, and enable more precise, patient-centered care. Specific applications include automated classification of neural activity, behavioral analysis in animal models, and multimodal data integration. A key concept introduced is “shadow …

Ai Institute Summer Camp Academic Preview Webinar: Curriculum, Research, And Outcomes, Paul English Applied Artificial Intelligence Institute

Paul English Applied Artificial Intelligence (AI) Institute Publications

This webinar presents an academic preview of the AI Institute Summer Camp hosted by the Paul English Applied Artificial Intelligence Institute at the University of Massachusetts Boston. The session introduces the program’s curriculum, structure, and student outcomes, providing insight into a hybrid learning model that combines faculty-led lectures, hands-on labs, and guided project development. The webinar highlights the program’s five-week structure, covering topics such as machine learning, neural networks, computer vision, speech and language processing, and generative AI. Participants learn how students engage in real-world AI applications, complete portfolio-ready projects, and develop research and presentation skills. This session is designed …

Robust Hardware Trojan Detection Leveraging Dual‑Domain Features And Stacked Ensemble Learning, Sefatun-Noor Puspa

All Theses

In Cyber-physical systems rely on sensors, communication, and computing, all powered by integrated circuits (ICs). These ICs are vulnerable to malicious hardware attacks, with hardware Trojans being one of the stealthiest threats. Trojans are malicious implants in the circuitry, which are often inserted during design or fabrication stages. This stealthy addition remains dormant until triggered and might cause functional disruptions or sensitive information leakage once triggered. Traditional IC validation methods, such as functional testing and logic analysis, usually fail to capture these subtle anomalies because hardware Trojans are intentionally designed to mimic normal circuit behavior. They often remain dormant under …

Ai-Based Framework For Elucidating The Ring Opening Energy Of Cycloalkanes, Evan Griffin

Open Access Theses & Dissertations

Cycloalkanes are a prominent component of jet fuels due to their high volumetric energy density and lower soot yield compared to aromatics. However, the structural nuance of cycloalkanes can make predicting their combustion kinetics challenging. The present study aims to develop a quantitative structure-property relationship of the ring opening energy (ROE) of cycloalkanes. A total of 253 cycloalkanes were curated from the ZINC15 database, ranging from monocycloalkanes to tetra-cycloalkanes. The 3-D cartesian coordinates of the closed shell and open shell diradical states for all curated cycloalkanes was generated with RDKit in Python and used MMFF94 to optimize the molecular geometries. …

A Low-Cost Motion Classification System For A Stuffed Animal Using An Imu And Machine Learning, Rachel N. Guynes

Honors Theses

One of the many fields that has seen the integration of robots is therapy. Zoomorphic robots (ZR) are designed to look and behave like animals to assist in Animal Assisted Therapy (AAT) practices. Studies show that ZRs can provide benefits similar to working with an actual animal; however, their high cost limits their accessibility. This thesis documents the process of building a real-time, low-cost motion classification system that can be attached to a stuffed animal to make it more interactive. Using a Random Forest (RF) classifier, the system identifies movements with approximately 81.67% accuracy.

Physics-Based And Data-Driven Low-Order Modeling Of Centrifugal Compressors, Chase Oliphant

Theses and Dissertations

Centrifugal compressors serve as critical components across many applications, yet their design and performance prediction present significant challenges. While computational fluid dynamics provides high-fidelity predictions of compressor behavior, its prohibitive computational cost renders it impractical for preliminary design phases where rapid exploration of the geometric and operating space is essential. Conversely, existing low-order models have enabled efficient performance estimation but suffer from limited accuracy and poor interpretability, particularly when applied to novel geometries or off-design operating conditions. This dissertation addresses these limitations through three complementary contributions that bridge the gap between high-fidelity CFD and preliminary design tools. The first contribution …

Predictive Analytics In Oncology And Ophthalmology: Machine Learning Applications For Diabetic Retinopathy And Breast Cancer, Ali Abidalkareem

Electronic Theses and Dissertations 2020 - Present

The convergence of artificial intelligence and healthcare represents one of the most transformative developments in modern medicine, with deep learning technologies emerging as powerful tools for addressing complex diagnostic challenges. This dissertation develops and validates machine learning frameworks that address critical challenges in medical diagnosis through innovative approaches to data augmentation, feature learning, and classification, focusing on two fundamental problems: Diabetic Retinopathy (DR) severity classification using multi-model convolutional neural networks (CNNs), and breast cancer stage identification using microRNA (miRNA) gene expression biomarkers. For diabetic retinopathy classification, this work proposes an ensemble deep learning framework that integrates Diffusion-based data augmentation for …

Ais26s: Genai And Llms For Cybersecurity, Political Sciences, Transportation Security, And Resiliency, Latifur Khan

Paul English Applied Artificial Intelligence (AI) Institute Publications

This presentation examines the role of generative artificial intelligence (GenAI) and large language models (LLMs) in addressing complex challenges across cybersecurity, political science, and transportation security. Dr. Latifur Khan discusses how advanced AI methods can be applied to threat detection, data analysis, and decision-making in high-risk and data-intensive environments. The talk highlights interdisciplinary applications of LLMs, emphasizing their ability to extract insights from large-scale data, improve system resilience, and support intelligent infrastructure. Emerging research directions and practical implications for real-world deployment are also discussed.

Traditional And Machine-Learning Equalization Techniques For Bandwidth-Limited Short-Reach Optical Communication Channels, Abdullah Khawatmi

Theses

This thesis investigates equalization techniques for bandwidth-limited short-reach optical communication systems, with a focus on Visible Light Communication (VLC) and Step-Index Plastic Optical Fiber (SI-POF) links. Commercial light-emitting diodes and photodiode receivers impose severe bandwidth constraints, inter-symbol interference, and noise sensitivity, which fundamentally limit achievable data rates. The work addresses these impairments through systematic evaluation of traditional digital signal processing–based equalizers and modern machine-learning-based post-equalization methods. The primary aim of this thesis is to enhance the achievable data rate and reliability of commercial short-reach optical links while maintaining practical computational complexity. Specifically, the objectives are to (i) design and experimentally …

Aiw26s: Machine Learning Of Structured Data, Moumita Saha

Paul English Applied Artificial Intelligence (AI) Institute Publications

This workshop introduces the fundamentals of machine learning for structured data, focusing on tabular datasets and real-world applications. Participants explore key concepts such as data types, data preprocessing, feature engineering, and supervised learning methods. The session covers commonly used models, including linear regression, logistic regression, decision trees, and neural networks, along with evaluation metrics such as RMSE, accuracy, and confusion matrices. By the end of the workshop, participants will have gained a practical understanding of how to build, interpret, and evaluate machine learning models for structured data.

Development Of Deep Fused Neural Architecture For Ancient Tamil Palm-Leaf Manuscript Recognition, Hariharan P Mr

Theses and Dissertations

Digitizing Tamil palm-leaf manuscripts is important for education, communication, and the preservation of cultural heritage. The complex structure of the Tamil script, the wide range of handwriting styles, and the degradation seen in ancient Tamil palm-leaf manuscripts make these texts very difficult to read and understand. Digital Image Processing (DIP), document analysis techniques, and traditional Optical Character Recognition (OCR) are unable to handle noise, background interference, faded ink, and limited labelled data, motivating the need for robust, effective Deep Learning (DL)- based solutions.

As a prerequisite to understanding and designing effective recognition systems for ancient manuscripts, this thesis first examines …

Quantification Of Interfacial Defects In Frp-Strengthened Rc Member Using Non-Destructive Testing Methods, Yuliya Tyrak

USF Tampa Graduate Theses and Dissertations

Externally bonded fiber-reinforced polymer (FRP) composites are widely used to strengthen reinforced concrete (RC) structures. However, the long-term performance of these systems depends on the integrity of the FRP–concrete bond. Interfacial defects such as debonding, delamination, cracking, and voids may reduce structural capacity and are often difficult to detect using conventional inspection techniques. This study investigates the effectiveness of nondestructive testing (NDT) methods for detecting interfacial defects in FRP-strengthened RC members. It explores the potential for automated defect identification using thermal imaging and machine learning.

Laboratory specimens with predefined defects were fabricated and tested using visual inspection (VI), tap testing …

Ai Applications In Assessing Risk For Periodontal Disease: A Systematic Review, Grant O. Korte, Claudia M. Tellez Freitas

Annual Research Symposium

Objectives:

The goal of this systematic review is to examine the impact of using artificial intelligence (AI) to predict a patient’s risk level for periodontal disease by analyzing proven systemic health diseases linked to periodontitis. This is being done by primarily focusing on prevention and early diagnosis using machine learning programs that have been proven effective in other fields of periodontitis research.

Methods:

To conduct this study, a comprehensive literature review of journals published after 2020 was performed through four databases: PubMed, Scopus, Web of Science and Dentistry and Oral Science Source. The search was completed in adherence …

Nlp-Based Comparative Assessment Of Clauses Quantifying Risk Transfer In Construction Contracts, Hoda M. Elshamy

Theses and Dissertations

Employers in the construction industry mostly deviate from standard contract forms such as FIDIC and NEC, by introducing alterations to the contract conditions that shift a great portion of risks from the client to the contractor. Originally, these risks were distributed more equitably between all contracting parties in the standards forms. The imbalances in the contractual conditions create fertile ground for conflicts, and if not resolved, will escalate to disputes during the project execution phase, leading to significant cost overruns and time delays. The contractors, therefore, attempt to restore the original balance through making amendments to the communicated contract, through …

Harnessing Ml And Iiot For Traceability In Continuous Production Systems: A Conceptual Framework, Kholoud M. Abdelaal

Theses and Dissertations

In the era of rapid technological advancement, the manufacturing sector faces increasing pressure to leverage emerging technologies to enhance operational efficiency and minimize waste. In this context, traceability plays a pivotal role, as it provides complete visibility of processes and products throughout manufacturing systems, enabling them to identify areas for improvement and take corrective actions accordingly. Additionally, traceability ensures compliance, supports product recalls, provides a clear understanding of the system’s performance, and enables fact-driven decision-making in multiple aspects of the manufacturing system. Although the broad spectrum of traceability applications in batch production-based plants, traceability remains challenging to achieve in continuous …

Computational Modeling And Machine Learning For The Design Of Polymer-Based Protective Systems Under Dynamic Loading, Jonathan Tate Villada

Open Access Dissertations

Computational modeling and machine learning offer powerful pathways for designing polymer-based protective systems subjected to dynamic loading, particularly for mitigating the early shock-dominated response generated by near-field underwater explosions (UNDEX). As naval, offshore, and submerged infrastructure systems continue to grow in strategic importance, there is an increasing need for lightweight, damage-tolerant protective solutions capable of reducing transmitted pressure, deformation, and energy transfer under extreme impulsive environments. Since large-scale experimental testing under such conditions are costly and limited, validated numerical frameworks provide an efficient means to evaluate polymeric coatings, architected metastructures, and data-driven predictive tools across broad design spaces. The first …