LOUIS University of Alabama in Huntsville
Not a member yet
8547 research outputs found
Sort by
A Simulation and Design Framework for Adapting Hydra 70 Rocket to GPS Aided Air to Ground Missile
Get PDFSolar Flare Forecasting Multiple ML and Curation Technique Study Hour-by-hour SHARP Parameter Data Archive
No full textHour-by-hour AR parameter data as a series of .csv files in a zip archive. Contains the filtered data described in the Solar Flare Forecasting using Machine Learning (ML) and SDO/HMI Data: Multiple ML Model and Data Curation Technique Comparison Study paper of Newman, Hall, Farris, Singh, Pogorelov, Benson, Raza, and Trital paper, 2025 submission date, for ApJS. Each file in the zip has data for one class of flares at a timepoint a certain number of hours in advance of flare onset. The first letter of such file names indicates flare class and the number before hrs in the title indicates the number of hours in advance of flare onset for the data in the file
Novel AI/ML-based frameworks for protein conformation selection in drug discovery applications
No full textDrug development is a lengthy, expensive process with a high failure rate. The urgency brought on by the COVID-19 pandemic has accelerated the integration of artificial intelligence (AI) and machine learning (ML) to enhance drug discovery by increasing target specificity, reducing toxicity, and optimizing formulation strategies. Building on this momentum, this dissertation introduces novel AI/ML-driven frameworks for protein conformation selection and classification, addressing critical challenges in modern drug discovery. However, designing such a novel AI/ML data-driven framework is a difficult task because most real-world biomedical datasets suffer from class imbalance issues, which can significantly skew AI/ML model training, resulting in biased models and poor prediction accuracy for the minority class. Another issue is the small sample sizes in biomedical datasets, which might complicate drug discovery by misclassifying drug candidate conformations. Hence, to address the aforementioned challenges, this dissertation presents a series of AI/ML data-driven methodologies that have the capability to work with smaller sample sizes suffering from class-imbalance while maintaining model performance. This dissertation presents multiple AI/ML data-driven frameworks aimed at: i) addressing the class imbalance issue in biomedical data, particularly in identifying potential binding protein conformations in a dataset where the non-binding protein conformation outnumbers the binding protein conformations, ii) using data-driven approaches to select probable physio-chemical features of potential binding protein conformations which could aid in identifying unique physio-chemical descriptors that could play a pivotal role in the binding capability of a protein conformation and also help in reducing the dimensionality of the dataset, allowing this work to be carried out on a personal computer rather than a supercomputer, iii) maximizing the prediction accuracy of binding and non-binding protein conformations, and iv) utilizing a Multi-modal Framework for Integrating Local and Global Descriptors via Graph Convolutional Networks. The AI/ML methodologies introduced are novel and innovative, striving to achieve a thorough understanding of the selection and prediction processes of binding protein conformations, which are crucial for drug discovery applications. The research outcomes from this work can help to streamline the development of new drugs and have a direct impact on the efficiency, cost-effectiveness, and speed with which novel therapeutic agents are introduced to the market
A comparative study of classical and deep feature learning methods for prediction of user request patterns for cloud resources
Get PDFLarge-scale cloud storage systems managing petabyte-scale data face cost challenges due to sparse, irregular file access patterns. Traditional attribute-based methods often fail to capture dynamic temporal and event-driven behaviors. This thesis compares classical statistical and deep learning approaches for predicting file access patterns using five months of real-world access logs from Tsinghua University’s FTP server (2.9 million events). Five forecasting models—ARIMA, SARIMA, Exponential Smoothing, Prophet, and 1D CNNs—are evaluated across 1,161 files. A novel hybrid clustering method combines time series similarity with forecasting accuracy metrics. Results show ARIMA significantly outperforms deep learning (2.3× better accuracy: 0.0188 vs. 0.036 MAE) for hourly forecasts, particularly for high-frequency files. Clustering achieves strong separation (silhouette score: 0.889), identifying four behavioral patterns that support targeted forecasting strategies with 40–60% improved accuracy. These insights enable cluster-specific intelligent data tiering, reducing storage costs by 30–50% while maintaining data accessibility
Characterization and comparison of skin and exoskeleton microbiomes of stygobiotic fauna and the influence of environmental variables on microbiome diversity and composition
No full textCaves and other subterranean ecosystems are among the most understudied ecosystems globally and house unique and ecologically important biodiversity. While several studies have characterized the microbial communities of respective cave systems, research on microbiomes of cave fauna has been limited. This study investigated microbiomes for aquatic cave fauna in the southern Cumberland Plateau of Tennessee and Alabama, USA. Exoskeleton and skin swabs were taken from three species of cave-dwelling crayfishes, two species of salamanders, and one cavefish. Bacterial operational taxonomic units (OTUs) identified in microbiomes of hosts included primarily members of the phylum Pseudomonadota, but also the phyla Acidobacteriota, Actinomycetota, Bacteroidota, Chloroflexota, Cyanobacteriota, and Planctomycetota. Environmental and land cover/land use (LULC) variables affected microbiome composition and diversity for hosts and the surrounding aquatic environment. Microbiome diversity and composition varied between sites primarily for Southern Cave Crayfish, and to some degree for Southern Cavefish and Tennessee Cave Salamanders
