Data Science Commons^™

Statistical Analysis Of Log Transformation Effectiveness In Air Traffic Movement Forecasting During Covid-19 In South Africa, John Lehlaka Masekoameng

Journal of Aviation Technology and Engineering

This study evaluates the effectiveness of log transformation in enhancing multiple regression models used to forecast air traffic movements (ATMs) in South Africa during the COVID-19 pandemic. Using 60 monthly observations from October 2016 to September 2021, the analysis incorporates variables such as revenue, lockdown levels, COVID-19 metrics, exchange rates, gross domestic product, and population. Two models are compared: one using raw ATMs and another with log-transformed ATMs as the dependent variable.

While the untransformed model shows stronger explanatory power (R² = 0.904, adjusted R² = 0.891) compared to the log-transformed model (R² = 0.772, adjusted R² = 0.741), the …

Topological Data Analysis Of New York City Taxi Trip Data Using Advanced Dimensionality Reduction And Clustering Techniques – Uncovering Structure And Timeliness, Mahalakshmi Sakthivel

Theses and Dissertations

In this project, we used Topological Data Analysis (TDA) to explore the shape and structure of high-dimensional data and the timeliness dimension of information quality through topological data analysis, with the long-term goal of automatically computing timeliness values that reflect how useful data items are for decision making. The project followed a two-phase approach: in the first half, we employed the Kepler Mapper library along with techniques like Principal Component Analysis (PCA), Uniform Manifold Approximation and Projection (UMAP), T-SNE (t-Distributed Stochastic Neighbor Embedding) and Customized Embedding to analyze and visualize complex datasets. In the second phase, we specifically applied our …

Reproducible Semantic Data Management Workflow For Materials Data Science: Generating Knowledge Graphs With Robust Fairifcation Pipelines, Kyle R. Henrikson, Van D. Tran, Meredith Francis, Isabella Giammattei, Quynh D. Tran, Laura S. Bruckman, Erika I. Barcelos, Roger H. French

Student Scholarship

Combining data from multiple sources is crucial for efficient knowledge aggregation in materials data science. FAIR data from ontology and Linked Data principles enable this. Semantic data management streamlines data exchange and aggregation, ensuring information is available and extractable. FAIRLinked and GraphDB provide solutions for consolidating, hosting, and extracting meaningful insight from multimodal data.

Data-Driven Prediction Of Superconducting Critical Temperature: A Linear And Regularized Linear Modeling Approach, Dipok Deb

Data Science and Data Mining

This study adopts a data-driven approach to estimate the critical temperature of superconducting materials using linear machine learning models. A comprehensive dataset derived from material physico-chemical properties was analyzed after systematic preprocessing and standardization. Three linear modeling strategies—Linear Regression, Ridge Regression, and Linear Regression with Subset Selection—were developed and evaluated using standard regression performance metrics. The findings demonstrate that both basic and regularized linear models can effectively capture the relationship between material features and superconducting behavior, offering robust and interpretable predictions. While feature selection enhances model transparency, it comes with a modest reduction in predictive capability. Overall, this work emphasizes …

Supercharging Simulation-Based Inference For Bayesian Optimal Experimental Design, Samuel Klein, Willie Neiswanger, Daniel Ratner, Michael Kagan, Sean Gasiorowski

Physics Faculty Publications

Bayesian optimal experimental design (BOED) seeks to maximize the expected information gain (EIG) of experiments. This requires a likelihood estimate, which in many settings is intractable. Simulation-based inference (SBI) provides powerful tools for this regime. However, existing work explicitly connecting SBI and BOED is restricted to a single contrastive EIG bound. We show that the EIG admits multiple formulations which can directly leverage modern SBI density estimators, encompassing neural posterior, likelihood, and ratio estimation. Building on this perspective, we define a novel EIG estimator using neural likelihood estimation. Further, we identify optimization as a key bottleneck of gradient based EIG …

Identifying Relevant Covariates In Rna-Seq Analysis By Pseudo-Variable Augmentation, Yet Nguyen, Dan Nettleton

Mathematics & Statistics Faculty Publications

RNA-sequencing (RNA-seq) technology allows for the identification of differentially expressed genes, which are genes whose mean transcript abundance levels vary across conditions. In practice, RNA-seq datasets often include covariates that are of primary interest in addition to a set of covariates that are subject to selection. Some of these covariates may be relevant to gene expression levels, while others may be irrelevant. Ignoring relevant covariates or attempting to adjust for the effect of irrelevant covariates can compromise the identification of differentially expressed genes. To address this issue, we propose a variable selection method that uses pseudo-variables to control the expected …

Computational Clinical Judgment: Predicting Risk With Large Language Models, Hannah Laqueur, Ryan W. Copus

Faculty Works

For seventy years, research has shown actuarial methods outperform clinical judgment. Yet actuarial approaches have limitations: they generally rely on structured data; cannot exploit rare case-specific details; have limited accuracy where outcome data are scarce or incomplete; and cannot offer case-level justifications. Large language models (LLMs) offer a different approach. Like actuarial methods, they aggregate information algorithmically, but like clinicians, they bring general knowledge and can provide case-level justifications. We prompted seven LLMs to assess rearrest risk from 113 parole hearing transcripts and compared their predictions to a machine learning model trained on 4,000 cases with 91 administrative variables. GPT-5 …

Cross-Temporal Statistical Approaches For Evaluating Predictor-Outcome Relationships, Jeff Joseph

Dartmouth College Ph.D Dissertations

Central auditory function is linked with cognitive deficits, but few research projects use existing statistical approaches or develop new ones to forecast cognitive deficits using the results of central auditory tests. To address this limitation, we use a series of statistical learning frameworks for predicting a child’s cognitive abilities based on his/her central auditory performances and demographic factors. Two key challenges exist. First, children may start the study at a time when they are unable to perform the central auditory tests or cognitive tests. Second, cognitive performance is age-dependent, particularly in the early formative years of childhood and adolescence. To …

Spatial Temporal Modeling Of Infectious Disease Patterns In Texas, Robert E. Lashbrook

Earth & Environmental Sciences Theses

The Texas Department of State Health Services monitors numerous notifiable conditions statewide, including Campylobacter, Salmonella, Shiga toxin-producing Escherichia coli (STEC), Rabies, and West Nile virus (WNV). Given the substantial health, economic, and public health burden associated with these conditions, improving prediction is an important step toward reducing their overall impact. This study evaluated whether external demographic, social, climate, and environmental data could improve prediction of county-year disease activity across Texas. County level data was analyzed using supervised machine learning models, including linear regression, ridge regression, multilayer perceptron, random forest, XGBoost, as well as K-means clustering to identify broader …

A Data-Driven Framework For Automation Readiness In Minnesota State University, Mankato Course Scheduling, Prisca Bongu Payanzo Maba

All Graduate Theses, Dissertations, and Other Capstone Projects

University course scheduling is one of the most complex optimization problems in higher education institutions. With universities growing in size and offering a broad spectrum of majors and disciplines, the number of possible course scheduling combinations increases exponentially, rendering traditional ways of scheduling ineffective.

Although operations research has extensively studied automated scheduling algorithms, there has been limited investigations into the organization readiness of academic departments to implement such systems. This paper offers a hybrid data science framework that assesses departmental readiness for scheduling automation.

The study combines qualitative Zoom interview data from 19 academic departments with institutional scheduling rules from …

An Association Test For Ordinal Outcomes In Clustered Data With Informative Cluster Size, Hasika K. Wickrama Senevirathne, Sandipan Dutta

Mathematics & Statistics Faculty Publications

In cluster-correlated data, the number of observations in a cluster can be associated with the outcome from that cluster. This phenomenon is known as informative cluster size which can occur in cluster-randomized clinical trial data. Several studies have found that ignoring the issue of informative cluster size can produce biased results in the analysis of clustered data. Most of the existing methods for addressing informative cluster size are suited to continuous outcomes. However, ordinal outcomes and covariates are often encountered in clustered data obtained from large clinical studies. The existing methods for ordinal association testing in clustered data can produce …

Logistic-T Multinomial Mixture Model For Clustering For Microbiome Data, Wenshu Dai, Yuan Fang, Sanjeena Subedi

Mathematics & Statistics Faculty Publications

The logistic-normal multinomial distribution has been used for modelling microbiome data obtained from high-throughput sequencing technologies, which are compositional in nature. A logistic-normal multinomial distribution is a hierarchical multinomial distribution that assumes the latent variable which are the additive log-ratio (ALR) transformed proportions in a multinomial distribution follows a Gaussian distribution. Model-based clustering algorithms have also been developed for clustering microbiome data based on the logistic-normal models. However, the Gaussian assumption may violated when the ALR transformed variable exhibit heavy-tailed distributions or has outliers. Our study introduces a novel mixture of logistic-t multinomial models that effectively address these challenges. Utilizing …

Validity Assessment Of Resting Heart Rate Variability From The Garmin Health Snapshot, Kayla M. Porter, Andrew Flatt

Honors College Theses

Purpose: To assess the agreement between the Garmin Forerunner 265 (a commercially available sports watch) and a single-channel electrocardiographic (ECG) chest strap for determining resting heart rate variability (HRV). Secondary aims were to assess the impact of skin tone and body position on measurement accuracy.

Methods: Young adults (n = 30, 57% women) aged 18–39 years without known cardiovascular conditions and without tattooing or scarring on the dorsal left wrist were recruited. HRV was recorded simultaneously using ECG and the Forerunner 265’s optical sensor during Garmin’s 2-minute “Health Snapshot.” Measurements were obtained in three standardized positions: supine, seated, and standing. …

Gradient-Based, Post-Optimality Sensitivity Analysis With Respect To Parameters Of State Equations, Gene Hou, Jonathan Degroff

Mechanical & Aerospace Engineering Faculty Publications

Design optimization is a computational tool that can enable a designer to investigate the effectiveness of a design concept in an organized format. However, this design process requires the design variables, constraints, and objective function to be properly defined and expressed in mathematical forms. Post-optimality analysis thus becomes a necessary step to investigate different variations in the problem formulation and parameters to ensure that optimization produces a stable and trustworthy outcome. One efficient way to achieve this aim is to compute the local derivative of the optimized objective function with respect to the optimization problem parameters, such as bounds on …

Mechanical-Medical Convergence In Heart Failure: Artificial Intelligence, Finite-Element Modeling, And 3d Printing For Diagnosis And Prognosis, Quazi Noor E. Sabrina, Quazi Md Zobaer Shah, Quazi Noor E. Sohela, Md Mahabub Hasan Mousum, Md. Moyeen Uddin Chisty, Quazi Md. Akbar Shah

Mechanical & Aerospace Engineering Faculty Publications

Heart failure remains a leading cause of global morbidity and mortality, yet routine clinical indices often miss the regional biomechanical disturbances that drive progression and shape treatment response. This State-of-the-Art review examines how finite-element (FE) modeling, additive manufacturing, and artificial intelligence (AI) are converging to improve the diagnosis, phenotyping, procedural planning, and prognostic assessment of heart failure (HF). Although these technologies have matured in structural heart disease and transcatheter intervention research, their greatest translational potential may lie in HF, where patient-specific ventricular remodeling, myocardial stress–strain heterogeneity, valve-ventricular coupling, and device-tissue interaction are incompletely captured by conventional clinical indices. We synthesize …

Systematic Approaches To Characterizing Vulnerabilities And Enhancing Robustness Of Text And Vision-Language Models, Poojitha Thota

Computer Science and Engineering Dissertations

The proliferation of artificial intelligence (AI) across critical domains, including news summarization, privacy-policy analysis, and medical decision support, has raised growing concerns about the security and robustness of these systems against adversarial manipulation. This dissertation investigates adversarial robustness in generative AI by addressing three key research goals: (1) characterizing adversarial vulnerabilities across generative models, (2) developing systematic defenses to improve the robustness of generative models, and (3) designing deployment-time safeguards for securing LLM interactions.

Towards the first goal, we characterize adversarial vulnerabilities across text-based and multimodal systems. In abstractive text summarization, we show that inference-time perturbations can exploit lead bias …

Previsit Ai: A Retrieval-Augmented Generation For Patient Readiness In Clinical Encounters, Rolande Umuhoza

All Graduate Theses, Dissertations, and Other Capstone Projects

With healthcare systems under growing pressure from rising patient volumes and shrinking consultation windows, improving how patients communicate with physicians has become essential to delivering quality care. Yet patients routinely arrive at appointments unable to clearly describe their symptoms, recall their medical history, or articulate concerns, contributing to miscommunication, diagnostic inefficiency, and pre-visit anxiety. This study introduces PreVisit AI, a conversational system designed to address this gap through structured, knowledge-based patient preparation. The system is built on a Retrieval-Augmented Generation (RAG) architecture combining HuggingFace sentence embeddings (all-MiniLM-L6-v2), a Chroma vector store, and Google’s Gemini language model over a curated seven-document …

Impact Of Cross-Client Heterogeneity In Federated Learning For Real-World Plant Disease Classification, Ritesh Janga

All Graduate Theses, Dissertations, and Other Capstone Projects

Deep learning applications are being adopted in agricultural image analysis that include challenges of data privacy and limited institutional data and heterogeneity of different types of architectures. However, Federated Learning is a model that allows collaborative training on data that does not have to be shared among parties. Therefore, Federated Learning is an effective method of collaborative training; however, its comparative effectiveness as compared to individual (local) training on diverse architectures has never been examined in an agricultural context. The objective of this study was to examine Federated Learning for the purpose of crop disease classification on extreme non-IID distributed …

Toward Interpretable Multi-Omics Multimodal Biomedical Artificial Intelligence, Yanjun Lyu

Computer Science and Engineering Dissertations

The complexity of human disease arises from biological processes that unfold across multiple scales, from molecular variation through cellular function, tissue organisation, brain phenotypes, each of which is associated with distinct measurement modalities, regularities, and characteristic. Contemporary biomedical artificial intelligence has brought the opportunity to reveal the complexity with in; however, its methodological default, in which models are trained on most readily available modality, does not adequately engage with the multi-scale connected structure by which biological meaning is constituted. The research area of multi-omics and multi-modal AI for biomedicine remains at an early exploratory stage, and the work presented in …

Predicting Criminal Behavior In Major Us Cities, Madison A. Price

SPARK Symposium Presentations

In recent years, especially post pandemic, there has been a decrease in crime in the United States. Unfortunately, the country’s violent crime rates are still significantly higher compared to similar high-income countries, so what predicts crime in major American cities? There is tons of research to support the idea that demographics can offer some insight into predicting crime. There are countless online resources that seek to identify major crime centrals in the United States (Petrino, 2025). In the late 1990s, researchers noticed that crime rates in cities had a downward slope due to an important contributor: demographic change (Fox & …