Data Science Commons^™

Dynamic Trust Calibration, Bruno Miranda Henrique

Dartmouth College Ph.D Dissertations

Trust calibration between humans and Artificial Intelligence (AI) is crucial for optimal decision-making in collaborative settings. Excessive trust can lead users to accept AI-generated outputs without question, overlooking critical flaws, while insufficient trust may result in disregarding valuable insights from AI systems, hindering performance. Despite its importance, there is currently no definitive and objective method for measuring trust calibration between humans and AI. Current approaches lack standardization and consistent metrics that can be broadly applied across various contexts, and they don’t distinguish between the formation of opinions and subsequent human decisions. This thesis brings a novel and objective method for …

Reimagining Less-Than-Truckload Pricing Development In Competitive Bid Environments With Artificial Intelligence, Lawson C. Levin

Data Science Undergraduate Honors Theses

This undergraduate thesis explores how data analytics and engineering judgment are used to support pricing decisions in the less-than-truckload (LTL) freight market. It’s based on an internship with ArcBest Corporation. It explains the company’s background, its role in the LTL market, and the responsibilities of a Pricing and Supply Chain Engineer within the Yield department.

Most of the internship was spent evaluating requests for proposals (RFPs), in which a negotiating third party provides a customer’s shipment data that must be cleaned, analyzed, and translated into a comprehensive pricing offer. Using the Data Science Analytics Process as a framework, this thesis …

Pyspqr: A Python Package For Density Estimation Using Deep Learning, Cameron Eddy, Reetam Majumder

Electrical Engineering and Computer Science Undergraduate Honors Theses

Splines are used for representing complex functions. In statistics, splines can be used for distributional shapes that are difficult to model by traditional parametric approaches. Ramsay (1) uses M-Spline bases to estimate continuous distributions. Semi-Parametric Quantile Regression (SPQR), developed by Xu and Reich (2), models conditional distributions where a neural network is used to estimate the basis function weights that depend on covariates. (3) implements a package for SPQR in R. We build on this by implementing a version of SPQR in Python with PyTorch. By using PyTorch, we can use more sophisticated deep learning architectures than those available in …

Recursion, Regurgitation, And Regeneration: Testing Limits And Revealing Biases Of Generative Ai Models Through Multimodal Feedback Loops, William Donnell-Lonon

Data Science Undergraduate Honors Theses

Contemporary generative AI systems such as OpenAI's GPT-4o and DALL-E models embed complex priors about society, reality, and history shaped by training data distributions, social alignment procedures, legal constraints, and safety regulations. This study uses a "telephone game" methodology to investigate how embedded social, political, and visual biases propagate and reveal themselves through iterative multimodal generation loops, where image captioning and text-to-image models are chained in successive feedback cycles.

Using CLIP similarity metrics, facial recognition algorithms, semantic drift analysis, and qualitative content observations, I tested how image subject matter affects the rate and quality of semantic and visual shift, identity …

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 …

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 …

Data Defines Success: Algorithm For Dataset Quality Assessment In Deep Learning For Malware Detection, Matei Ionescu

Theses and Dissertations

The field of artificial intelligence is based upon the premise of constructing architectures through which to propagate training data. However, the majority of existing research literature is focused on architecture. While necessary, the attention devoted to the architecture should not so precipitously exceed that of the data. It should be noted that this disparity is not without reasonable cause. Data quality is often exceedingly difficult to verify due to particularities of the field or subfield; LLM repositories of text are distinct from image recognition pictures of dog breeds which are distinct from EEG waveforms of human brains which are distinct …

Visualizing And Evaluating Binary Classifier Performance With Contingency Space, Colin D. Kehoe, Azim Ahmadzadeh

Undergraduate Research Symposium

Traditional metrics for evaluating binary classifiers, such as Accuracy, F1 Score, and True Skill Statistic (TSS), often obscure the underlying tradeoffs between true positive and true negative performance—particularly in imbalanced or high-stakes domains. This poster introduces the Contingency Space, a two-dimensional representation of classifier behavior defined by true positive rate (TPR) and true negative rate (TNR). Within this space, scalar performance metrics become geometric surfaces, revealing how scores vary across the entire landscape of possible classifier outputs.

We present a Python package that implements this framework, enabling users to map model predictions into the Contingency Space, visualize metric surfaces …

Face Value: A Computational Approach To Subjective Impressions Of Faces, Kevin Kpankou

Undergraduate Research Symposium

Various computational models of first impressions have been developed to uncover the mechanisms driving these judgments. However, the implicit notion of a singular ``human'' often overlooks meaningful individual differences in beliefs, attitudes, and associations, as well as culturally grounded group-level constructs. In this paper, we extend Cultural Consensus Theory (CCT) to estimate culturally shared beliefs about faces by incorporating latent constructs structured around interpretable facial features extracted via computer vision algorithms. We apply our model to a large-scale dataset of people’s first impressions of faces. Our approach reveals a robust mapping between facial features and culturally constructed impressions, allowing us …

Parameter Informed Reinforcement Learning For Vehicle System Identification, Nathan Schaff

Doctoral Dissertations and Master's Theses

Accurate system identification is essential for modeling and controlling vehicle dynamics. This dissertation explores the application of Parameter Informed Reinforcement Learning (PIRL) as a novel approach to system identification (SYSID). PIRL integrates prior system knowledge, such as physical parameters, into reinforcement learning (RL) frameworks to improve estimation accuracy. The study begins with an overview of traditional SYSID methods and then introduces PIRL as a modification of standard RL. The research applies PIRL to short-period aircraft dynamics, demonstrating its effectiveness in both offline and online learning frameworks. The dissertation then further explores PIRL’s utility in an indirect model reference adaptive control …

A Predictive Model For Multi- Week Respiratory Risk From Red Tide On Florida’S Gulf Coast., Elmer S. Ochaeta

Computer Science and Engineering Faculty Publications

Florida’s Gulf Coast red tide (Karenia brevis) can put toxins into the air, making people cough, irritating the throat, and worsening asthma or other breathing problems especially when winds blow from the ocean toward the beach. Right now, most public updates don’t really help with the question people actually ask when planning a weekend or vacation: “Will going to or close to the beach be risky in the next few weeks?”.

In this project, I build a weekly early warning system that estimates respiratory risk for specific beaches and predicts that risk 2 to 4 weeks ahead. The study covers …

Privacy Preserving-Based Artificial Intelligence For Precision Agriculture, Partha P. Sengupta

Dissertations

The research work finds a solution to precision agriculture of cotton cultivation using artificial intelligence (AI) models. Two sets of model performance based on the application are selected namely a low resource and a high resource setting. This is because using drone surveys to capture images identifying the classes of stressed and unstressed cotton plantation requires limited model architecture and CPU based computation. Thus, traditional AI models were selected for low resource settings. Again, for high computation intensive models like transfer learning-convolution neural network (CNN) based architectures were grouped into high resource settings. There was another issue of class imbalance …

Mapping The Research Landscape Of Covid-19 And Artificial Intelligence Using The Lens Database, Aichetou Mohamed Sidiya, Jailan Fouad Aljizawi

Effat Undergraduate Research Journal

Due to the rapid spread of the COVID-19 pandemic, all countries faced a significant challenge in their efforts to monitor and halt the spread of the virus. Moreover, researchers around the world raced into publishing to get to understand the effects of this pandemic on every aspect of our life whether that was economic, medical, or social. Thus, this research seeks to assess the quality of research on the application of Artificial Intelligence in the Covid-19 pandemic with an emphasis on Saudi Arabia. The research methodology is based on Bibliometric Analysis techniques and VosViewer for various bibliometric visualizations based on …

Robustness Investigation, Detection, And Defense Of Deep Learning Models Against False Data Injection, Amirhossein Nazeri

All Dissertations

This dissertation addresses the critical challenge of adversarial robustness in deep learning systems, focusing on two fundamental domains: time-series prediction and object detection. As these AI systems become increasingly deployed in safety-critical applications from power grid management to autonomous vehicles their vulnerability to adversarial attacks poses significant risks to infrastructure and human safety.

The first contribution introduces a novel stealthy black-box False Data Injection (FDI) attack specifically designed for quasi-periodic time-series data. Unlike existing attacks that produce easily detectable anomalies, our method generates adversarial perturbations that preserve the underlying periodicity and statistical properties of the data, effectively bypassing traditional anomaly …

Ai Project Facilitation Guidance For Research Computing And Data (Rcd) Professionals, Anna Alber, Laura Briggs, Paul Brunk, Manasvita Joshi, Atish P. Kamble, Amira Kefi, Timothy Middelkoop, Semir Sarajlic, Ana Maria Sokovic, Jeffrey N. Valdez, Ying Zhang

Administration and Staff Articles and Research

The role of Artificial Intelligence (AI) in research and education continues to rapidly grow, resulting in increased collaboration between researchers in AI and Research Computing and Data (RCD) professionals to meet the research and teaching demands. RCD professionals bridge the gap between research and technology by guiding and collaborating with researchers and educators through the process of selecting the hardware, software, and services best suited for executing their AI projects. This includes ensuring compliance with funding and regulatory requirements across the entire lifecycle of the project. In this paper, we present an overview of the AI project lifecycle and how …

Property Testing Ai: An Efficient Frontier, Paul Sopher Lintilhac

Dartmouth College Ph.D Dissertations

In this dissertation, we take a step towards addressing the major problem of a lack of standardized and rigorous approaches to testing and evaluation of AI systems. Taking inspiration from both the fields of Property Testing and Property Based Testing (for programs), we develop a novel taxonomy of partially overlapping classes of properties of AI systems, including simple properties, compound properties, higher order properties, data relation properties, and architecture-utility properties. We argue that this taxonomy categorizes a diverse set of AI traits -- including accuracy, fairness, robustness, monotonicity, point-wise and global privacy properties, sensitivity, and more -- according to the …

Computer Vision In Soccer: Yolov11 Analytics Engine For Quantifying Game Strategy, Connor S. Maurer

Data Science Undergraduate Honors Theses

Single-shot object detection capabilities significantly reduce computational overhead for real-time computer vision in sports analytics at 60 FPS. YOLO11’s lightweight CNN gives promising accuracy while meeting the low-latency demand of dynamic soccer matches. As data-driven approaches take over the sport of soccer, efficient player tracking systems become critical for informing coach’s strategies. I prototype the ETL (Extract, Transform, Load) process of data collected from a single- shot detection program and evaluate its viability for estimating player fatigue. YOLO11 detects players, the ball, and other characteristics, with the output transformed by homography to estimate the positions in the real world. These …

Toward Robust Semantic Segmentation In Levee Infrastructure Monitoring: Enhancing Accuracy With High-Fidelity Synthetic Data And Ensemble Learning, Padam Jung Thapa

LSU New Orleans Theses and Dissertations

Abstract: Levees serve as critical flood protection structures, but failures due to inadequate maintenance and extreme water pressures have led to devastating events such as Hurricane Katrina. Manual inspections are slow, labor-intensive, and prone to human error, necessitating the development of automated solutions. This study proposes an AI-driven framework for levee inspection utilizing deep learning-based semantic segmentation to detect rutting and enhance the identification of sand boils. To address dataset limitations, high-fidelity synthetic images are generated using DreamBooth for fine-tuning, while ControlNet adds structural constraints to enhance realism and consistency. A semi-automatic convex hull annotation technique enhances labeling efficiency, and …

Extending Feature-Based Detection For Artificial Intelligence, Kayla Ahrndt

SPARK Symposium Presentations

AI text generation is rapidly developing, and, as a result, it is becoming increasingly difficult to differentiate it from human written text. Our base study by Leon Fröhling et al. proposed a feature-based detection model trained on GPT2, GPT3, and Grover data, as well as human-generated text. Our work extends their research by training a modified model with four neural networks on word embeddings, select features from the original study, as well as updated data (GPT3, GPT4, and Grover).

A Reinforcement Learning Approach For Maneuvering And Firing Decisions In Sead Operations, Nathaniel Garcia

Theses and Dissertations

The integration of automated processes in defense continues to expand, enhancing the lethality of military forces. Artificial intelligence accelerates decision-making cycles, removes the constraints of human-operated hardware, and improves coordination by enabling seamless integration across multiple systems. Suppression of Enemy Air Defenses (SEAD) missions are critical to the United States (U.S.) military, as they neutralize hostile air defense systems, ensuring air superiority and enabling safe and effective operations for aircraft in contested environments. Therefore, it is necessary to pair emerging autonomous capabilities with an important mission set in defense. This research investigates the Autonomous Unmanned Air-to-Ground Strike (AUAGS) problem, modeling …

Data Injustice In Global Justice, Asaf Lubin, Cherry Tang

Articles by Maurer Faculty

In May 2020, the United Nations Secretary-General unveiled a sweeping “Data Strategy for Action by Everyone, Everywhere,” seeking to unlock the UN’s “full data potential.” The International Criminal Court’s Office of the Prosecutor followed suit, declaring in 2023 its intent to acquire advanced cyber forensic tools so as to hold the “widest range of digital evidence globally.” Across international institutions, data-driven governance has become the norm, with humanitarian agencies and tribunals transforming into “data hubs and information clearinghouses.” This Article critiques the unfettered datafication of global justice by international courts and organizations. These entities have aggressively expanded their data-driven operations …

Streamlining The Data Mining Process Through Ai-Driven Prompt Templates, Mia Montevirgen, Drew Yan, Clara Lu, Laurent Shen, Weihong Ni

Capstone Showcase

With the increasing use and relevancy of AI in the world, this project aims to harness the power of AI, specifically ChatGPT, to streamline the process of data mining workflows. By developing custom prompt templates, this project seeks to utilize OpenAI API to assist with key data mining tasks, including data understanding, importing, and cleaning. This approach aims to increase workflow speed, reproducibility, and accessibility in data mining projects. The effectiveness of these prompt templates is evaluated by applying them to diverse datasets and assessing their impact on accuracy, efficiency, and reproducibility. Overall, the project highlights the potential to use …

Rethinking Retrieval Augmented Fine-Tuning In An Evolving Llm Landscape, Nicholas Sager, Timothy Cabaza, Matthew Cusack, Ryan Bass, Joaquin Dominguez

SMU Data Science Review

This study explores the utilization of Retrieval Augmented Fine-Tuning (RAFT) to enhance the performance of Large Language Models (LLMs) in domain-specific Retrieval Augmented Generation (RAG) tasks. By integrating domain-specific information during the retrieval process, RAG aims to reduce hallucination and improve the accuracy of LLM outputs. We investigate the use of RAFT, an approach that enhances LLMs by incorporating domain-specific knowledge and effectively handling distractor documents. This paper validates previous work, which found that RAFT can considerably improve the performance of Llama2-7B in specific domains. We also expand upon previous work into new state-of-the-art open-source models and other datasets with …

Exploring Healthcare Chatbot Information Presentation: Applying Hierarchical Bayesian Regression And Inductive Thematic Analysis In A Mixed Methods Study, Samuel Nelson Koscelny

All Theses

High blood pressure, also known as hypertension, significantly increases the risk of heart disease and stroke, which are leading causes of death in the United States. While contributing to over 691,000 deaths in 2021 alone in the United States (U.S.), it also imposes immense economic burden on the healthcare system, costing approximately $131 billion annually. One way to address this issue is for increased self-care behaviors and medication adherence, both of which require sufficient health literacy. Despite the importance of health literacy, 90% of U.S. adults struggle with health-related subjects. Overcoming the issues associated with health literacy requires addressing the …

Artificial Intelligence In Fusion Protein Three-Dimensional Structure Prediction: Review And Perspective, Himansu Kumar, Pora Kim

Faculty, Staff and Student Publications

Recent advancements in artificial intelligence (AI) have accelerated the prediction of unknown protein structures. However, accurately predicting the three-dimensional (3D) structures of fusion proteins remains a difficult task because the current AI-based protein structure predictions are focused on the WT proteins rather than on the newly fused proteins in nature. Following the central dogma of biology, fusion proteins are translated from fusion transcripts, which are made by transcribing the fusion genes between two different loci through the chromosomal rearrangements in cancer. Accurately predicting the 3D structures of fusion proteins is important for understanding the functional roles and mechanisms of action …

Toward The Integration Of Behavioral Sensing And Artificial Intelligence, Subigya K. Nepal

Dartmouth College Ph.D Dissertations

The integration of behavioral sensing and Artificial Intelligence (AI) has increasingly proven invaluable across various domains, offering profound insights into human behavior, enhancing mental health monitoring, and optimizing workplace productivity. This thesis presents five pivotal studies that employ smartphone, wearable, and laptop-based sensing to explore and push the boundaries of what these technologies can achieve in real-world settings. This body of work explores the innovative and practical applications of AI and behavioral sensing to capture and analyze data for diverse purposes. The first part of the thesis comprises longitudinal studies on behavioral sensing, providing a detailed, long-term view of how …

Code For Care: Hypertension Prediction In Women Aged 18-39 Years, Kruti Sheth

Electronic Theses, Projects, and Dissertations

The longstanding prevalence of hypertension, often undiagnosed, poses significant risks of severe chronic and cardiovascular complications if left untreated. This study investigated the causes and underlying risks of hypertension in females aged between 18-39 years. The research questions were: (Q1.) What factors affect the occurrence of hypertension in females aged 18-39 years? (Q2.) What machine learning algorithms are suited for effectively predicting hypertension? (Q3.) How can SHAP values be leveraged to analyze the factors from model outputs? The findings are: (Q1.) Performing Feature selection using binary classification Logistic regression algorithm reveals an array of 30 most influential factors at an …

Gender Detection In Facial Images: A Comprehensive Cnn Analysis, Jose N T Ambrosio, Anas Hourani, Magdalene Moy

SACAD: Scholarly Activities

This research investigates the construction of a robust gender detection system using facial features and Convolutional Neural Networks (CNNs), exploring the impact of different layer configurations on accuracy and computational efficiency. With a validation accuracy of 91%, findings illuminate the nuanced relationship between precision and computational resources, enriching discussions on facial recognition technologies.

Artificial Intelligence Could Probably Write This Essay Better Than Me, Claire Martino

Augustana Center for the Study of Ethics Essay Contest

No abstract provided.

Leveraging Explainable Artificial Intelligence To Optimize Clinical Decision Support, Siru Liu, Allison B Mccoy, Josh F Peterson, Thomas A Lasko, Dean F Sittig, Scott D Nelson, Jennifer Andrews, Lorraine Patterson, Cheryl M Cobb, David Mulherin, Colleen T Morton, Adam Wright

Faculty, Staff and Student Publications

OBJECTIVE: To develop and evaluate a data-driven process to generate suggestions for improving alert criteria using explainable artificial intelligence (XAI) approaches.

METHODS: We extracted data on alerts generated from January 1, 2019 to December 31, 2020, at Vanderbilt University Medical Center. We developed machine learning models to predict user responses to alerts. We applied XAI techniques to generate global explanations and local explanations. We evaluated the generated suggestions by comparing with alert's historical change logs and stakeholder interviews. Suggestions that either matched (or partially matched) changes already made to the alert or were considered clinically correct were classified as helpful. …