Data Science Commons^™

An Analytical Framework For Quantifying Urban And Community Resilience To Natural Hazards From Cell-Phone Gps-Location And Traffic-Flow Data, Georgios Chatzikyriakidis

Civil and Environmental Engineering Theses and Dissertations

Urban areas are increasingly exposed to natural hazards while accommodating a growing share of the global population, yet a consistent science-based framework for quantifying urban and community resilience remains lacking. This dissertation develops a physics-based analytical framework grounded in statistical mechanics and the quantitative theory of Brownian motion. A city is conceptualized as a complex medium in which citizens move analogously to Brownian particles within a viscoelastic environment, influenced by socioeconomic interactions and infrastructure functionality.

A central premise is that urban resilience, interpreted as engineering resilience (an outcome), can be quantified through a single metric: the mean-square displacement MSD=⟨r²(t)⟩, of …

Addressing The Problems Of Data Variations, Quality, And Scarcity In Training Deep Neural Networks, Jian Sun

Electronic Theses and Dissertations

The performance of deep neural networks (DNNs) is strongly influenced by the characteristics and quality of the underlying datasets. This Ph.D. dissertation addresses three pervasive data challenges-imbalance, quality degradation, and scarcity-that commonly hinder the effectiveness of DNNs in computer vision (CV) and natural language processing (NLP) applications.

Class imbalance remains one of the most frequent causes of degraded model generalization. While Focal Loss effectively mitigates inter-class imbalance by assigning higher weights to minority classes, it struggles with intra-class imbalance, particularly in video datasets where longer clips dominate feature representation. To address this, I implement and utilize …

Utilizing Computer Modeling To Optimize Electric Fields Within Xenon Time Projection Chambers, Miles Meloni

Honors Theses

XENONnT is a physics experiment designed with the goal of detecting dark matter particles. The detector is a time projection chamber; a series of charged electrodes creates an electric field, surrounding a central body filled with liquid and gaseous xenon. Photomultiplier tubes (PMTs), positioned on either end of the chamber, serve to detect light signals. We seek to minimize the root mean square of the electric field norms experienced by the PMTs. This quantity corresponds to the variance in the electric field observed by the PMTs. Establishing a consistent electric field is important to maintaining these sensitive components. The electric …

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 …

Hybrid Graph-Recurrent Architecture For Citation Recommendation Via Future Embedding Forecasting, Mohammad Ausaf Ali Haqqani

Computer Science and Engineering Theses and Dissertations

The rapid expansion of scientific literature has intensified the challenge of identifying relevant citations, particularly for newly published or under-cited papers. Traditional citation recommendation systems typically model static relationships or respond to past citation activity, offering limited predictive power for emerging works. In response, this thesis presents a temporal modeling framework for citation recommendation that anticipates future scholarly relevance by forecasting the latent representations of academic papers.

Building on prior work that utilized Temporal Graph Networks (TGNs) to model dynamic citation flows, we propose Graph-Time, a hybrid architecture that integrates a Graph Transformer with a GRU-based time series predictor. The …

Ml Playground: Data Modification/Preprocessing And Model Simulation Tool, Marco D. Cerrato

Electronic Theses, Projects, and Dissertations

There is a heavy reliance on programming when it comes to learning machine learning (ML). This often creates barriers for students and newcomers unfamiliar with coding. While the lessons you learn in the classroom provide essential foundational understanding, some technical or practical aspects of ML—such as data preprocessing, feature engineering, and model tuning—are best learned through hands-on interaction. ML Playground was developed to act as a proof-of-concept application to address this gap by offering a browser-based, graphical user interface that lets users engage with core ML workflows without writing code. Designed with educational accessibility in mind, the application allows users …

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).

Improving The Completeness Of Food Composition Databases Using Predictive Analysis., Carla Arenhart

ICT

This study investigates the use of machine learning regression models to impute missing micronutrient values in Food Composition Databases (FCDBs), focusing on the FAO/INFOODS dataset. A cascading prediction methodology leverages nutrient interdependencies to systematically estimate missing values. Four models—Random Forest (RF), Support Vector Machine (SVM), Gradient Boosting Machines (GBM), and Deep Neural Networks (DNN)—were evaluated using MAE, MSE, RMSE, and R². RF and GBM achieved the highest predictive accuracy for protein, phosphorus, calcium, and magnesium, demonstrating that ML-based predictive analytics can provide a more reliable alternative to traditional imputation methods. These findings support improved dietary assessments, nutritional research, and data-driven …

Implementation Of Time Series And Neural Networks For Forecasting Agricultural Prices In The Irish Market: A Comparative Analysis Of Milk, Beef, And Potatoes., César Augusto Núñez

ICT

Agricultural price volatility represents a central challenge for the Irish agri-food sector, affecting the stability of producers, cooperatives, and policymakers. This study aimed to compare three predictive approaches applied to strategic commodities such as milk, beef, and potatoes: a traditional statistical time series model (SARIMA) and two deep learning architectures (RNN and LSTM). Using historical price series collected over a decade, the models were developed and evaluated following a rigorous methodological process that included data preparation, algorithm training, and validation of results using performance metrics widely used in time series research. The findings show that the SARIMA model was most …

Improving Fairness In Convolutional Neural Networks For Demographic Face Classification., Leandro Andrade

ICT

This study examines racial bias mitigation in Convolutional Neural Networks (CNNs) for demographic face classification using the FairFace dataset. Three architectures—ResNet50, VGG19, and InceptionV3—are evaluated, with dataset balancing strategies including undersampling and class weighting. Results indicate that InceptionV3 with class weighting achieves the most consistent performance across racial groups, with improved F1-scores and generalization through hyperparameter optimization and data augmentation. Challenges remain in distinguishing visually similar groups, highlighting the need for equitable datasets and fairness-aware training. These insights are critical for ensuring accuracy and fairness in applications such as law enforcement, healthcare, and human–computer interaction.

Traditional Vs Deep Learning Approaches For Efficient Electricity Consumption Prediction., Lucas Sant’Ana

ICT

Accurate forecasting of electricity demand is critical for reliable energy planning, resource allocation, and policy design. Traditional statistical models, such as ARIMA, SARIMA, and ARIMAX, have been widely applied but remain constrained by linear assumptions, limited temporal memory, and inflexibility in handling multiple exogenous drivers. In this study, we systematically compare these approaches with multivariate Long Short-Term Memory (LSTM) networks, which are capable of capturing nonlinear dependencies, long-term temporal dynamics, and multivariate interactions. Historical electricity consumption data were combined with weather variables, including temperature, wind speed, and rainfall, and pre-processed through cleaning, scaling, and temporal alignment. Statistical baselines and deep …

Time Series Forecasting In Financial Markets: Benchmarking The Temporal Fusion Transformer Against N-Beats., Fergus Fleury

ICT

This study compares the performance of two deep learning architectures, the Temporal Fusion Transformer (TFT) and N-BEATS, for 10-day stock price forecasting. Both models were implemented using the Darts Python library, which ensured consistent preprocessing, training, and evaluation. The dataset, sourced from Yahoo Finance, included daily equity prices, technical indicators, a market sentiment index, and earnings announcements.

TFT was applied as a multivariate model incorporating past, future, and static covariates, while N-BEATS was trained as separate univariate models with past covariates only. A rolling forecast cross-validation approach was used for evaluation. Results show that TFT consistently outperformed N-BEATS, particularly under …

Tour Demand Forecasting In Ireland: Development And Evaluation Of Classical, Deep Learning, And Hybrid Models, Ruben Elias Charleston Montfort

ICT

Tourism plays a significant role in global economies by supporting employment, infrastructure, and national development. As international travel continues to grow, accurate tourism demand forecasting has become increasingly important for effective planning and decision-making. In Ireland, tourism is a key economic sector attracting millions of visitors annually. For tour operators such as Irish Day Tours, reliable demand forecasting is essential for optimizing logistics, resource allocation, marketing strategies, and customer satisfaction. Advances in machine learning and deep learning techniques offer new opportunities to improve forecasting accuracy and support data-driven decision-making within the tourism industry.

Comparative Evaluation Of Ai-Generated Synthetic Data And Real-World Data Performance In Predictive Analytics., Corey Louise Hughes

ICT

There are growing restraints when it comes to Real World Data (RWD), these include topics such as privacy regulations, ethical concerns, and the cost of collecting the data, and they have drove an interest in AI-generated synthetic data as a potential alternative in predictive analytics. This project examines the possibilities of synthetic data and if it can act as a reliable substitute for RWD in predictive modelling. This project uses Wasserstein Generative Adversarial Networks with Gradient Penalty (WGAN-GP) to generate synthetic reproductive health data and evaluates its predictive performance against RWD using linear regression and key metrics, including Mean Absolute …

Enhancing Uk Electricity Price Forecasting Using Deep Learning., Stephen Cooke

ICT

Accurate short-term electricity price forecasting (EPF) is crucial for efficient operation of the UK’s multi-layered power market, impacting generators, traders, the ESO, and policymakers. Prices are highly volatile and non-linear due to renewables, demand fluctuations, and market coupling across Day-Ahead, Intraday, and Balancing Mechanism venues. Traditional statistical models often fail under such dynamics, while machine learning and deep learning approaches—particularly LSTM, GRU, and hybrid architectures—effectively capture temporal dependencies and exogenous drivers. Empirical evidence shows that these models outperform classical baselines, enabling more accurate scheduling, risk management, and financial savings.

Identifying And Forecasting Key Drivers Of Greenhouse Gas Emissions In Ireland's Residential Sector Multivariate Time Series Analysis., Sallam Noor Aldeen Salman

ICT

This study evaluates advanced time-series forecasting models to predict greenhouse gas emissions (GHGE) in Ireland's residential sector. LSTM, XGBoost, and ARIMA models were tested alongside feature selection methods including PCA, XGBoost-based importance, and Granger causality. Urban population growth and electricity consumption emerged as the most significant predictors. While LSTM struggled due to limited data, XGBoost showed strong predictive performance (MAPE ~9%), and ARIMA with key features achieved the highest accuracy (MAPE 3.78%). Forecasts indicate a declining GHGE trend, offering actionable insights for environmental planning and policy in the residential sector.

Understanding Model Behaviour And Interpretability In Time Series Forecasting: A Deep Dive Into Lstm And Gru With Xai Techniques., Federico Ariton

ICT

Time-series forecasting is widely used in data analytics, yet the interpretability of deep learning models remains a key challenge. This study compares Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) models using a dual interpretability framework that combines attention mechanisms and SHAP analysis. Two feature sets were evaluated across assets with different volatility regimes: Solana and Shiba Inu (high volatility), Bitcoin (moderate volatility), and Apple (low volatility). Results show that autoregressive features achieved the lowest forecasting errors, while volatility- and momentum-based indicators provided stronger interpretability. GRU performed best in moderate-volatility conditions, whereas LSTM demonstrated more consistent performance across varying …

Brain Tumor Classification Using Deep Learning- Poster, Rayen Bentemessek

ICT

The project presents deep learning solutions to classify brain tumors through MRI images. Two Convolutional Neural Network (CNN) models were developed, a custom CNN designed from scratch and a pretrained ResNet50 that was transfer learned and fine-tuned.

Both models were implemented following CRISP-DM methodology from data understanding to deployment, and they were evaluated using different metrics such as accuracy, precision, recall and F1-score.

Key Highlights: •The custom CNN model achieved higher accuracy but failed to locate tumors. •ResNet50 provided a good performance while balancing explainability through Grad-CAM. •Model was deployed through Gradio to demonstrate a real-world use of the solution.

European Air Pollution And The Proposed Timelines Of Implementing The World Health Organization 2021 Air Quality Guidelines Ca3., Lukia Hartin

ICT

This research examines Ireland’s air pollution trends and evaluates whether current reductions in PM2.5, PM10, and NO₂ are sufficient to meet the WHO 2021 Air Quality Guidelines by 2040. Using four years of EPA-validated pollutant data (2020–2023), alongside Building Energy Rating (BER) and national transport datasets, the study applies the CRISP-DM methodology to guide analysis, preprocessing, modelling, and evaluation. Extensive data cleaning and alignment were required due to inconsistent station coverage, varying formats, and missing values. Forecasting models—including Random Forest, Gradient Boosting, SVR, and linear regression—were assessed using MSE, R², and trend significance to project pollutant levels across different Irish …

Customer Service Support. Utilizing Machine Learning To Classify, Prioritize And Summarize Issues., Hoai Nhan Nguyen

ICT

This capstone project investigates the application of machine learning and natural language processing (NLP) to enhance customer support operations through automated ticket classification, prioritization, and summarization. Using the multilingual Customer Support Emails dataset from Kaggle, the project follows the CRISP-DM methodology, performing extensive data cleaning, preprocessing, feature engineering, and class balancing. Five machine learning models—Decision Tree, KNN, LinearSVC, Naive Bayes, and Random Forest—were evaluated using hyperparameter tuning, cross-validation, confusion matrix analysis, and learning curves. LinearSVC demonstrated the strongest performance for both queue and priority classification, achieving accuracies of 89.8% and 81.2% respectively, with consistent generalization across folds. For summarization, extractive …

Player Transfer Market In European Football Using Machine Learning To Analyse The Evolution Of The European Football., Pablo Lopes De Souza Oliveira

ICT

The study uses machine learning to analyse the European football transfer market from 2015 to 2025, revealing patterns in transfer fees and market values influenced by player attributes, highlighting the potential of data-driven insights.

Predicting Early Customer Inactivity In The Banking Sector Using Machine Learning: A Churn Prevention, Ivana Mc Fadden

ICT

Customer churn, when customers stop using a company’s services, is a challenge for the banking sector (Singh et al., 2023). High churn rates often signal poor customer experiences, resulting in revenue losses and increased costs to obtain new clients. Goyal and Srivastava (2015) stress that fostering loyalty through exceptional service and understanding customer needs is important for long-term retention.

This project aims to predict early customer inactivity, an indicator of churn, by using machine learning. Early identification of at-risk customers will allow banks to apply targeted interventions, reduce acquisition costs, and improve customer satisfaction (Singh et al., 2023). By analysing …

Neural Networks Activation Functions And Hybrid Activations Functions Accuracy And Loss Comparison On Small Dataset Against Large Datasets For Classification Problems, Antonio Felipe Cora Martins

ICT

Even on the era of Big Data, small datasets are the reality of many companies and sectors. Many datasets in rare disease diagnosis, custom manufacturing, military sciences, bioengineering, and disaster events are commonly limited in size, making machine learning predictive modelling difficult. Being the Activation Function choice crucial for Neural Networks learning, it raises the question of their effectiveness in such scenarios. This study compares five standard (single) activation functions (Sigmoid, Tanh, ReLU, Leaky ReLU, ELU) and two hybrid variants (one a mix of ReLU plus Tanh and a Learnable Activation Function with a trainable weight (alpha) that balances ReLU …

Big Data Vs Big Law: The Impact Of Big Data And Machine Learning In Anonymising Or Synthesizing Data For Use Across Borders., Kenneth Darker

ICT

This research investigates the viability of anonymization and synthetic data generation in the area of big data so that the data could be shared across borders and exist outside the constraints of privacy laws. These privacy laws are growing around the world to help protect individual identity and prevent open sharing of private data. These privacy laws all provide guidance on how data may be shared and the strict conditions upon how that may occur. Two methods which are growing in popularity are anonymization of data, specifically k-Anonymity, l-Diversity and t-Closeness, and generating synthetic data from a real dataset leveraging …

The Actuarial Applications Of Machine Learning And Big Data In The Life Assurance Industry: Managing Customer Retention And Customer Outcomes By The Application Of Data Science., Brian Cunningham

ICT

Lapses are an issue in the insurance industry in general. They affect a company’s profitability, cash flows and solvency. High levels of lapses can cause reputational damage that could provoke a cycle of even more lapses. It is therefore incumbent on a company to do its utmost to retain the business it has written for the term it was written for.

If a company could predict which of its policies were about to lapse, it could proactively attempt to prevent them by contacting the policyholder and engaging in a discussion to ascertain the likelihood of their choosing to leave. In …

Assemble The Ensemble: A Multi Model Approach For Customer Churn Prediction In The Gambling Industry., Paul Corcoran

ICT

Churn rates are remarkably high in the gambling industry, an extremely competitive landscape coupled with a severe lack of brand loyalty among its customer base makes churn prediction one of the main problems an operator will face. This paper explores the range of possible modelling solutions with a key emphasis on ensemble learning to improve on existing methods. During this exploration, a host of modelling techniques are formulated with a focus on scalability facilitated by Apache Spark distributed computing language. Thirteen variations of models, including single classifiers and ensemble families are evaluated as to their suitability in solving the problem. …

The Use Of Deep Learning Solutions To Develop A Practice Tool To Support Lámh Language For Communication Partners, Gabriel Bueno Pimentel Borges

ICT

This study has proposed an alternative to promote the learning and enhancement of Lámh language for communication partners that support current users by creating a real time detection tool to recognise 20 chosen Lámh signs based on existing studies in the field. This implementation was carried out by generating primary data composed by MediaPipe landmark numpy arrays of 40 frames and 45 repetitions per sign. The Neural Networks were built using the Python library Keras and the applied SVM models were built with the library sklearn. The real time detection was carried out by integrating the mentioned elements with the …

Premier League Results Predictions, Laura Consuegra

ICT

This project applies machine learning techniques to predict outcomes in the English Premier League, one of the most prestigious and widely followed football competitions worldwide. By analysing historical and real-time match data, including performance metrics such as goals scored, shots on target, and cards received, predictive models are developed to forecast match results with higher accuracy. The study evaluates the effectiveness of these models and explores the influence of key statistical features on team performance. The findings aim to provide strategic insights for fans, bookmakers, and coaching staff, supporting performance evaluation, tactical decision-making, and a deeper understanding of the factors …

A Comparative Analysis Of Machine Learning And Neural Network Performance In House Price Prediction: Dublin Vs. Other Irish Regions, Diarmuid Carroll

ICT

This research presents a comparative analysis of machine learning and neural network performance in predicting house prices across Ireland’s regional housing markets. It addresses important methodological challenges and offers new empirical insights into how market complexity influences algorithm accuracy and performance. Drawing on 627,294 residential transactions from the Irish Property Price Register (2012–2024), the study applies a dual validation strategy, temporal and stratified sampling, across four regional classifications: Dublin, Other Cities, the Commuter Belt, and Rural areas.

The study makes three main contributions to data analytics theory and practice. First, it identifies and resolves temporal confounding effects in algorithm evaluation. …

Predictive Analytics For Customer Churns In Financial Services., Thant Thiha

ICT

This project presents a customer churn prediction analysis in the telecommunications sector, achieving an ROC-AUC of approximately 0.86 using statistically validated features and interpretable AI models. Key churn drivers identified include the number of products held, customer age, and geographic location. Ensemble models, such as Random Forest and Gradient Boosting, provided the highest predictive performance. Ethical AI principles were applied to ensure fairness, transparency, privacy, and accountability. Business insights derived from the analysis inform targeted retention strategies, prioritising multi-product users, specific age groups, and geographic segments. Deployment recommendations include the tuned Random Forest model with ongoing monitoring, governance, and future …