Data Science Commons^™

Learning From Non-Stationary Data Streams, Gabriel Jonas Aguiar

Theses and Dissertations

The rapid growth of data from sources such as mobile applications, sensors, and network monitoring has increased the need for machine learning algorithms capable of handling non-stationary data streams. However, learning from such streams presents significant challenges due to their evolving nature and the presence of concept drift. One of the most complex issues is learning from imbalanced data streams, where shifting data distributions, combined with feature space drifts, complicate continuous adaptation. These challenges become even more pronounced in multi-class scenarios, which are common in real-world applications. Detecting concept drift in such contexts is particularly demanding, as it requires tracking …

Check Your Data Before You Wreck Your Model: The Impact Of Careless Responding On Substance Use Data Quality, Abby L. Braitman, Anna M. Petrey, Jennifer L. Shipley, Rachel Ayala Guzman, Emily Renzoni, Alison Looby, Adrian J. Bravo

Psychology Faculty Publications

Background: The accuracy of survey responses is a concern in research data quality, especially in college student samples. However, examination of the impact of removing participants from analyses who respond inaccurately or carelessly is warranted given the potential for loss of information or sample diversity. This study aimed to understand if careless responding varies across a number of demographic indices, substance use behaviors, and the timing of survey completion.

Method: College students (N = 5809; 70.7% female; 75.7% White, non-Hispanic) enrolled in psychology classes from six universities completed an online survey assessing a variety of demographic and substance use-related information, …

Landslide At River's Edge: Alum Bluff, Apalachicola River, Florida, Joann Mossa, Yin-Hsuen Chen

Center for Geospatial Science, Education & Analytics Faculty Publications

When rivers impinge on the steep bluffs of valley walls, dynamic changes stem from a combination of fluvial and mass wasting processes. This study identifies the geomorphic changes, drivers, and timing of a landslide adjacent to the Apalachicola River at Alum Bluff, the tallest natural geological exposure in Florida at similar to 40 m, comprising horizontal sediments of mixed lithology. We used hydrographic surveys from 1960 and 2010, two sets of LiDAR from 2007 and 2018, historical aerial, drone, and ground photography, and satellite imagery to interpret changes at this bluff and river bottom. Evidence of slope failure includes a …

Computational Bridges: Enhancing Natural Language Processing Of Swahili., Joyce Murungi

Harrisburg University Other Works

Swahili remains significantly underrepresented in natural language processing (NLP) despite being one of the most widely spoken languages in Africa. Computational Bridges: Enhancing Natural Language Processing of Swahili addresses this gap through computational linguistics, corpus creation, and large-scale analysis of Swahili syntax and lexical structure. Central to this study is GUMZO, a novel corpus developed from spontaneous conversational data collected from YouTube videos, television panel discussions, political speeches, religious discourse, and unscripted broadcasts. Unlike many existing datasets that rely on formal or translated text, GUMZO captures authentic language use and provides a stronger foundation for NLP research involving low-resource languages. …

Safeguard Cyberspace In Ransomware Era: Risk Analysis & Cyber Insurance, Li Huang

Electronic Theses & Dissertations (2024 - present)

The increasing frequency and severity of ransomware attacks pose significant challenges for organizational cybersecurity. Fragmentation across disciplines in cyber defense has created practical gaps in the development of the necessary capabilities needed to address rapidly evolving cyber threats. This study explores the impact of ransomware attacks and the evolving role of cyber insurance as a proactive cybersecurity partner. Bridging the gap between actuarial science and cyber risk management, it proposes an interdisciplinary framework that quantifies the impact of ransomware and integrates cyber insurance into cybersecurity strategies.

The primary contribution of this study is methodology. We present a framework that remains …

Empirical Vulnerability Function Development Based On The Damage Caused By The 2014 Chiang Rai Earthquake, Thailand, Patcharavadee Hong, Masashi Matsuoka

Institute for Innovation & Entrepreneurship Publications

Seismic hazards in Thailand are frequently overlooked in disaster management planning, leading to insufficient research and significant economic losses during earthquake events. The 2014 Chiang Rai earthquake exposed critical vulnerabilities in Thailand's building practices due to widespread non-compliance with building codes and limited preparedness. This exposure prompted the development of empirical vulnerability functions using loss data from 15,031 damaged residences. The study analyzed government compensation records, which were standardized using replacement cost metrics. Three distinct models were developed through probabilistic and possibilistic modeling approaches. Residual analysis demonstrated the superior performance of the possibilistic approach, with the Possibilistic-based Vulnerability Function achieving …

Fares On Fairness: Using A Total Error Framework To Examine The Role Of Measurement And Representation In Training Data On Model Fairness And Bias, Patrick Oliver Schenk, Christoph Kern, Trent D. Buskirk

Data Science Faculty Publications

Data-driven decisions, often based on predictions from machine learning (ML) models are becoming ubiquitous. For these decisions to be just, the underlying ML models must be fair, i.e., work equally well for all parts of the population such as groups defined by gender or age. What are the logical next steps if, however, a trained model is accurate but not fair? How can we guide the whole data pipeline such that we avoid training unfair models based on inadequate data, recognizing possible sources of unfairness early on? How can the concepts of data-based sources of unfairness that exist in the …

Training Set Augmentation And Harmonization Enables Radiomic Models To Detect Early Onset Of Lung Cancer, Claire Huchthausen, Menglin Shi, Gabriel L.A. Sousa, James Larner, Einsley Janowski, Jonathan Colen, Krishni Wijesooriya

Data Science Faculty Publications

Radiomics-based machine learning models have the potential to detect lung cancer at inception from CT scans and transform patient outcomes. Low malignancy rates in early-development pulmonary nodules (PNs) and variable image acquisition hinder development of clinically applicable radiomics-based early detection models. To address these challenges, we augmented training using later-development PNs and harmonized for acquisition effects. We first trained machine learning models to predict PN malignancy using radiomic features from scans of early-development benign and malignant PNs (n = 187) harmonized using ComBat. Observing near-chance performance, we augmented training with later-development benign and malignant PNs (n = 225). We evaluated …

An Overview Of The Special Issue, Rachel Gibson, Trent Buskirk

Data Science Faculty Publications

[Introduction] Across the quantitative social sciences, researchers increasingly face significant challenges and opportunities prompted by the arrival of new sources of very rich, highly granular, and often unstructured digital data. While traditional methods such as surveys and content analysis tools remain indispensable for measuring individual attitudes, behaviors, demographic characteristics, and media messaging online, they often struggle to capture the complex multimodal information streams and metadata generated by social media platforms, mobile devices, sensors, and tracking applications. Collecting and analyzing these diverse new forms of content, dynamic moment-to-moment behaviors, and naturally occurring interactions has become a pressing and exciting research task-one …

Heterogeneous Clustering Of Multiomics Data For Breast Cancer Subgroup Classification And Detection, Joseph Pateras, Musaddiq Lodi, Pratip Rana, Preetam Ghosh

Computer Science Faculty Publications

The rapid growth of diverse -omics datasets has made multiomics data integration crucial in cancer research. This study adapts the expectation–maximization routine for the joint latent variable modeling of multiomics patient profiles. By combining this approach with traditional biological feature selection methods, this study optimizes latent distribution, enabling efficient patient clustering from well-studied cancer types with reduced computational expense. The proposed optimization subroutines enhance survival analysis and improve runtime performance. This article presents a framework for distinguishing cancer subtypes and identifying potential biomarkers for breast cancer. Key insights into individual subtype expression and function were obtained through differentially expressed gene …

A Data-Driven Sliding-Window Pairwise Comparative Approach For The Estimation Of Transmission Fitness Of Sars-Cov-2 Variants And The Construction Of The Evolution Fitness Landscape, Md Jubair Pantho, Richard Annan, Landen Alexander Bauder, Sophia Huang, Letu Qingge, Hong Qin

Computer Science Faculty Publications

Estimating the transmission fitness of SARS-CoV-2 variants and understanding their evolutionary fitness trends are important for epidemiological forecasting. Existing methods are often constrained by their parametric natures and do not satisfactorily align with the observations during COVID-19. Here, we introduce a sliding-window data-driven pairwise comparison method, the differential population growth rate (DPGR) that uses viral strains as internal controls to mitigate sampling biases. DPGR is applicable in time windows in which the logarithmic ratio of two variant subpopulations is approximately linear. We apply DPGR to genomic surveillance data and focus on variants of concern (VOCs) in multiple countries and regions. …

Benchmarking Batch-Effect Correction Methods Towards The Construction Of A Triple-Negative Breast Cancer Cell Atlas, Peter Scheible, Amy H. Tang, Jing He, Jiangwen Sun

Computer Science Faculty Publications

Triple-negative breast cancer (TNBC) requires detailed cellular mapping given its aggressive nature, immense tumor heterogeneity and genetic diversity. We integrated 156,794 cells from six scRNA-seq datasets—including tumors, metastases, and cell lines—to build a TNBC scRNA cell atlas, focusing on batch effect mitigation while maintaining biological and molecular details. Preprocessing f ilters noise, normalizes data, and leverages PCA for integration readiness. We utilized scANVI, a semi-supervised tool, to align datasets, preserving TNBC’s complex tumor heterogeneity via marker annotations [1]. UMAPs demonstrate biological clustering in integrated data, contrasted with datasetdriven unintegrated patterns. Assessments verifying effective batch correction. This method aligns with NASA’s …

Ai For Nuclear Physics: The Exclaim Project, S. Liuti, D. Adams, M. Boër, G. W. Chern, M. Cuic, M. Engelhardt, G. R. Goldstein, B. Kriesten, Y. Li, H. W. Lin, M. Sievert, D. Sivers

Computer Science Faculty Publications

An overview of the recent activity of the newly funded EXCLusives with AI and Machine learning (EXCLAIM) collaboration is presented. The main goal of the collaboration is to develop a framework to implement AI and machine learning techniques in problems emerging from the phenomenology of high energy exclusive scattering processes from nucleons and nuclei, maximizing the information that can be extracted from various sets of experimental data, while implementing theoretical constraints from lattice QCD. A specific perspective embraced by EXCLAIM is to use the methods of theoretical physics to understand the working of ML, beyond its standardized applications to physics …

Decode The Workload: Training Deep Learning Models For Efficient Compute Cluster Representation, Ahmed Hossam Mohammed, Mark Jones, Diana Mcspadden, Malachi Schram, Bryan Hess, Kishansingh Rajput

Computer Science Faculty Publications

In this study, we address the mounting challenge of monitoring high throughput computing clusters running computationally intensive jobs, which increasingly strains system administrators. We develop autoencoders that analyze traces of Linux kernel CPU metrics to capture salient system features by producing robust compressed embeddings for various downstream tasks. In addition, we employ graph neural networks to incorporate contextual information from surrounding CPUs and assess their performance. We also demonstrate the enhanced job differentiation achieved by increasing the sampling rate of these traces. Our models are evaluated based on their ability to generate meaningful latent representations, detect anomalies, and distinguish between …

Neural Topic Modeling Via Contextual And Graph Information Fusion, Jiyuan Liu, Jiaxing Yan, Chunjiang Zhu, Xingyu Liu, Qing Li, Yanghui Rao

Computer Science Faculty Publications

Topic modeling is a powerful unsupervised tool for knowledge discovery. However, existing work struggles with generating limited-quality topics that are uninformative and incoherent, which hindering interpretable insights from managing textual data. In this paper, we improve the original variational autoencoder framework by incorporating contextual and graph information to address the above issues. First, the encoder utilizes topic fusion techniques to combine contextual and bag-of-words information well, and meanwhile exploits the constraints of topic alignment and topic sharpening to generate informative topics. Second, we develop a simple word co-occurrence graph information fusion strategy that efficiently increases topic coherence. On three benchmark …

S²Il: Structurally Stable Incremental Learning, S. Balasubramanian, P. Yedu Krishna, Talasu Sai Sriram, M. Sai Subramaniam, Manepalli Pranav Phanindra Sai, Ravi Mukkamala

Computer Science Faculty Publications

Feature Distillation (FD) strategies are proven to be effective in mitigating Catastrophic Forgetting (CF) seen in Class Incremental Learning (CIL). However, current FD approaches enforce strict alignment of feature magnitudes and directions across incremental steps, limiting the model’s ability to adapt to new knowledge. In this paper, we propose Structurally Stable Incremental Learning (S²IL), a FD method for CIL that mitigates forgetting by focusing on preserving the overall spatial patterns of features which promote flexible (plasticity) yet stable representations that preserve old knowledge (stability). We also demonstrate that our proposed method S²IL achieves strong incremental accuracy and outperforms other FD …

Effective Pii Extraction From Llms Through Augmented Few-Shot Learning, Shuai Cheng, Shu Meng, Haitao Xu, Haoran Zhang, Shuai Hao, Chuan Yue, Wenrui Ma, Meng Han, Fang Zhang, Zhao Li

Computer Science Faculty Publications

Large Language Models (LLMs) exhibit strong natural language processing capabilities but also pose significant privacy risks, particularly regarding the leakage of Personally Identifiable Information (PII) embedded in their training data. Existing PII extraction methods suffer from the limitations of low success rates or impracticality for large-scale PII extraction. In this study, we propose a novel PII extraction approach based on enhanced few-shot learning techniques, which achieves efficient and cost-effective PII retrieval without relying on fine-tuning or jailbreaking. We evaluated our approach on both open-source and closed-source LLMs. The experimental results demonstrate that, for non-targeted PII extraction, the attack success rate …

A Universal Implementation Of Radiative Effects In Neutrino Event Generators, Júlia Tena-Vidal, Adi Ashkkenazi, Lawrence B. Weinstein, Peter Blunden, Steven Dytman, Noah Steinberg

Physics Faculty Publications

Due to the similarities between electron-nucleus (eA) and neutrino-nucleus scattering (νA), eA data can contribute key information to improve cross-section modeling in eA and hence in νA event generators. However, to compare data and generated events, either the data must be radiatively corrected or radiative effects need to be included in the event generators. We implemented a universal radiative corrections program that can be used with all reaction mechanisms and any eA event generator. Our program includes real photon radiation by the incident and scattered electrons, and virtual photon exchange and photon vacuum polarization diagrams. It …

A Method For Empirically Assessing Small Area Estimators Via Bootstrap-Weighted K-Nearest-Neighbor Artificial Populations, With Applications To Forest Inventory, Grayson W. White, Jerzy Wieczorek, Zachariah W. Cody, Emily X. Tan, Jacqueline O. Chistolini, Kelly S. Mcconville, Tracey S. Frescino, Gretchen G. Moisen

Faculty Journal Articles

National Forest Inventories monitor forest attributes across a variety of spatial and temporal scales in a given country. Increased interest in reporting and management at smaller scales has driven National Forest Inventories to investigate and adopt small area estimation (SAE) due to the promise of increased precision at these scales. However, comparing and evaluating SAE models for a given application is inherently difficult. Typically, many areas lack enough data to check unit-level modeling assumptions or to assess unit-level predictions empirically; and no ground truth is available for checking area-level estimates. Design-based simulation from artificial populations can help with each of …

Depletion Of Adipose Stroma-Like Cancer-Associated Fibroblasts Potentiates Pancreatic Cancer Immunotherapy, Zhongming Zhao, Mikhail G Kolonin

Faculty, Staff and Student Publications

This study shows that populations of CAFs have distinct effects on pancreatic cancer progression and shows that depletion of CAFs expressing adipose markers potentiates tumor/metastasis suppression effects of immune checkpoint blockade.