Digital Commons Network^™

Voice Synthesis Improvement By Machine Learning Of Natural Prosody, Joseph Kane, Michael N. Johnstone, Patryk Szewczyk

Research outputs 2022 to 2026

Since the advent of modern computing, researchers have striven to make the human–computer interface (HCI) as seamless as possible. Progress has been made on various fronts, e.g., the desktop metaphor (interface design) and natural language processing (input). One area receiving attention recently is voice activation and its corollary, computer-generated speech. Despite decades of research and development, most computer-generated voices remain easily identifiable as non-human. Prosody in speech has two primary components—intonation and rhythm—both often lacking in computer-generated voices. This research aims to enhance computer-generated text-to-speech algorithms by incorporating melodic and prosodic elements of human speech. This study explores a novel …

A Machine Learning Approach For Multipath Characterization And Mitigation Using Chipshape Observations, Sean A. L. Quiterio

Theses and Dissertations

Multipath continues to be a significant error source in satellite navigation. Recent solutions with Neural Networks (NN) model the effects of multipath on the autocorrelation function to predict errors in the Delay Lock Loop (DLL). Chipshape correlation provides a detailed look into the spreading code transitions in the time domain. It is useful in applications such as Signal Quality Monitoring (SQM) and is much more sensitive to multipath than autocorrelation. This research proposes NN models that each predict pseudorange or carrier range errors due to multipath by monitoring the chipshape correlation output. For a simulation with 50 MHz precorrelation bandwidth …

U.S. Army Cadet Command Branch Prediction Model, Daniel M. Krizan

Theses and Dissertations

The current system for providing US Army ROTC cadets their branches leaves significant uncertainty until the final pronouncement of branch assigned. This uncertainty can be alleviated by providing a prediction model for cadets to input personal data and desired branch to identify likelihood of receiving the request. This thesis produces a machine learning model capable of producing branch prediction for cadets.

Federated Analysis Of Wearables Data For United States Air Force Mental And Physical Readiness, Sung Yong O

Theses and Dissertations

This thesis investigates the use of machine learning and deep learning models within a federated learning framework to predict physical and mental readiness in military personnel, using wearable technology data. The collaboration with the 711th Human Performance Wing’s STRONG Lab highlights the importance of readiness as emphasized by the National Defense and Security Strategies. The study evaluates various predictive models, incorporating federated learning to ensure data privacy and security in healthcare systems. By analyzing a comprehensive dataset, the research aims to contribute to military readiness enhancement through technological advancements, supporting health and wellness initiatives to bolster the effectiveness of military …

Artificial Intelligence And Perception: An Empirical Study, Anthony J. Neil

Theses and Dissertations

This thesis investigates the impact of adjusting artificial intelligence explainability levels’ outputs on user perception. The overarching study extends within the Explainable Artificial Intelligence (XAI) domain. It examines a spectrum of variables, including performance, cognizance, familiarity, transparency, system bias, and the overall impact of AI, to understand their collective and individual effects that enable effective professional use in an organization. The study aims to illuminate the relationship between the degree of explainability provided by large language models such as ChatGPT, Bard, and Bing AI and the performance of these models when tasked with XAI adjustments.

Sensor-Based Vehicle Classification Using Machine Learning, Luke Mcfadden

Theses and Dissertations

This research investigates the classification of vehicles into heavy and light categories using acoustic, seismic, and magnetic sensor data. The effectiveness of using frequency domain data and classical machine learning techniques, is compared with the effectiveness of using time-series data and neural networks. The primary aim in doing so was to understand if modern neural network architectures could effectively remove the need for more traditional frequency based signals processing. A significant deliverable of this thesis was the feature importance determined for each of the three phenomenological types found within the data (acoustic, seismic, and magnetic). By analyzing the importance of …

The Impact Of Artificial Intelligence And Machine Learning On Organizations Cybersecurity, Mustafa Abdulhussein

Doctoral Dissertations and Projects

As internet technology proliferate in volume and complexity, the ever-evolving landscape of malicious cyberattacks presents unprecedented security risks in cyberspace. Cybersecurity challenges have been further exacerbated by the continuous growth in the prevalence and sophistication of cyber-attacks. These threats have the capacity to disrupt business operations, erase critical data, and inflict reputational damage, constituting an existential threat to businesses, critical services, and infrastructure. The escalating threat is further compounded by the malicious use of artificial intelligence (AI) and machine learning (ML), which have increasingly become tools in the cybercriminal arsenal. In this dynamic landscape, the emergence of offensive AI introduces …

Predicting Forage Provision Of Grasslands Across Climate Zones By Hyperspectral Measurements, F. A. Männer, J. Muro, J. Ferner, S. Schmidtlein, A. Linstädter

IGC Proceedings (1977-2023)

The potential of grasslands’ fodder production is a crucial management measure, while its quantification is still laborious and costly. Remote sensing technologies, such as hyperspectral field measurements, enable fast and non-destructive estimation. However, such methods are still limited in transferability to other locations or climatic conditions. With this study, we aim to predict forage nutritive value, quantity, and energy yield from hyperspectral canopy reflections of grasslands across three climate zones. We took hyperspectral measurements with a field spectrometer from grassland canopies in temperate, tropical and semi-arid grasslands, and analyzed corresponding biomass samples for their quantity (BM), metabolizable energy content (ME) …

Cancergpt For Few Shot Drug Pair Synergy Prediction Using Large Pretrained Language Models, Tianhao Li, Sandesh Shetty, Advaith Kamath, Ajay Jaiswal, Xiaoqian Jiang, Ying Ding, Yejin Kim

Faculty, Staff and Student Publications

Large language models (LLMs) have been shown to have significant potential in few-shot learning across various fields, even with minimal training data. However, their ability to generalize to unseen tasks in more complex fields, such as biology and medicine has yet to be fully evaluated. LLMs can offer a promising alternative approach for biological inference, particularly in cases where structured data and sample size are limited, by extracting prior knowledge from text corpora. Here we report our proposed few-shot learning approach, which uses LLMs to predict the synergy of drug pairs in rare tissues that lack structured data and features. …

Predicting Open-Pit Mine Production Using Machine Learning Techniques, Faustin Nartey Kumah, Alex Kwasi Saim, Millicent Nkrumah Oppong, Clement Kweku Arthur

Journal of Sustainable Mining

In mining, where production is affected by several factors, including equipment availability, it is necessary to develop reliable models to accurately predict mine production to improve operational efficiency. Hence, in this study, four (4) machine learning algorithms – namely: artificial neural network (ANN), random forest (RF), gradient boosting regression (GBR) and decision tree (DT)) – were implemented to predict mine production. Multiple Linear Regression (MLR) analysis was used as a baseline study for comparison purposes. In that regard, one hundred and twenty-six (126) datasets from an open-pit gold mine were used. The developed models were evaluated and compared using the …

Strategies To Combine 3d Vasculature And Brain Cta With Deep Neural Networks: Application To Lvo, Uma M Lal-Trehan Estrada, Arnau Oliver, Sunil A Sheth, Xavier Lladó, Luca Giancardo

Faculty, Staff and Student Publications

Automated tools to detect large vessel occlusion (LVO) in acute ischemic stroke patients using brain computed tomography angiography (CTA) have been shown to reduce the time for treatment, leading to better clinical outcomes. There is a lot of information in a single CTA and deep learning models do not have an obvious way of being conditioned on areas most relevant for LVO detection, i.e., the vasculature structure. In this work, we compare and contrast strategies to make convolutional neural networks focus on the vasculature without discarding context information of the brain parenchyma and propose an attention-inspired strategy to encourage this. …

Implications Of Lithium-Ion Cell Temperature Estimation Methods For Intelligent Battery Management And Fast Charging Systems, Ehab Bayoumi

Mechanical Engineering

This article examines in depth the most recent thermal testing techniques for lithium-ion batteries (LIBs). Temperature estimation circuits can be divided into six divisions based on modeling and calculation methods, including electrochemical computational modeling, equivalent electric circuit modeling (EECM), machine learning (ML), digital analysis, direct impedance measurement, and magnetic nanoparticles as a base. Complexity, accuracy, and computational cost-based EECM circuits are feasible. The accuracy, usability, and adaptability of diagrams produced using ML have the potential to be very high. However, both cannot anticipate low-cost integrated BMS live due to their high computational costs. An appropriate solution might be a hybrid …

Forecasting The Strength Of Graphene Nanoparticles-Reinforced Cementitious Composites Using Ensemble Learning Algorithms, Majid Khan, Roz-Ud-Din Nassar, Waqar Anwar, Mazhar Rasheed, Taoufik Najeh, Yaser Gamil, Furqan Farooq

Faculty Publications

Contemporary infrastructure requires structural elements with enhanced mechanical strength and durability. Integrating nanomaterials into concrete is a promising solution to improve concrete strength and durability. However, the intricacies of such nanoscale cementitious composites are highly complex. Traditional regression models encounter limitations in capturing these intricate compositions to provide accurate and reliable estimations. This study focuses on developing robust prediction models for the compressive strength (CS) of graphene nanoparticle-reinforced cementitious composites (GrNCC) through machine learning (ML) algorithms. Three ML models, bagging regressor (BR), decision tree (DT), and AdaBoost regressor (AR), were employed to predict CS based on a comprehensive dataset of …

Forecasting The Strength Of Graphene Nanoparticles-Reinforced Cementitious Composites Using Ensemble Learning Algorithms, Majid Khan, Roz-Ud-Din Nassar, Waqar Anwar, Mazhar Rasheed, Taoufik Najeh, Yaser Gamil, Furqan Farooq

Faculty Publications

Contemporary infrastructure requires structural elements with enhanced mechanical strength and durability. Integrating nanomaterials into concrete is a promising solution to improve concrete strength and durability. However, the intricacies of such nanoscale cementitious composites are highly complex. Traditional regression models encounter limitations in capturing these intricate compositions to provide accurate and reliable estimations. This study focuses on developing robust prediction models for the compressive strength (CS) of graphene nanoparticle-reinforced cementitious composites (GrNCC) through machine learning (ML) algorithms. Three ML models, bagging regressor (BR), decision tree (DT), and AdaBoost regressor (AR), were employed to predict CS based on a comprehensive dataset of …

Longitudinal Single-Cell Transcriptional Dynamics Throughout Neurodegeneration In Sca1, Leon Tejwani, Neal G Ravindra, Changwoo Lee, Yubao Cheng, Billy Nguyen, Kimberly Luttik, Luhan Ni, Shupei Zhang, Logan M Morrison, John Gionco, Yangfei Xiang, Jennifer Yoon, Hannah Ro, Fatema Haidery, Rosalie M Grijalva, Eunwoo Bae, Kristen Kim, Regina T Martuscello, Harry T Orr, Huda Y Zoghbi, Hayley S Mcloughlin, Laura P W Ranum, Vikram G Shakkottai, Phyllis L Faust, Siyuan Wang, David Van Dijk, Janghoo Lim

Duncan NRI Faculty and Staff Publications

Neurodegeneration is a protracted process involving progressive changes in myriad cell types that ultimately results in the death of vulnerable neuronal populations. To dissect how individual cell types within a heterogeneous tissue contribute to the pathogenesis and progression of a neurodegenerative disorder, we performed longitudinal single-nucleus RNA sequencing of mouse and human spinocerebellar ataxia type 1 (SCA1) cerebellar tissue, establishing continuous dynamic trajectories of each cell population. Importantly, we defined the precise transcriptional changes that precede loss of Purkinje cells and, for the first time, identified robust early transcriptional dysregulation in unipolar brush cells and oligodendroglia. Finally, we applied a …

Bitcoin Short-Term Price Prediction Using Time Series Analysis, Alanood Alkamali

Theses

This thesis explores the application of Autoregressive Integrated Moving Average (ARIMA) model to predict Bitcoin prices, a prominent and volatile cryptocurrency. The research falls within the context of financial forecasting, focusing specifically on the cryptocurrency markets. The primary research question is: “Can time series analysis be used to predict the future price of Bitcoin?” To answer this question, historical Bitcoin daily price data from 17/09/2014 to 17/09/2023 was obtained from Kaggle and analyzed. The study employs ARIMA modeling techniques to capture the autocorrelation, seasonality, and trend present in Bitcoin price time series. As a prerequisite for ARIMA modeling, the data …

Dataset Of Arabic Spam And Ham Tweets, Sanaa Kaddoura, Safaa Henno

All Works

This data article provides a dataset of 132421 posts and their corresponding information collected from Twitter social media. The data has two classes, ham or spam, where ham indicates non-spam clean tweets. The main target of this dataset is to study a way to classify whether a post is a spam or not automatically. The data is in Arabic language only, which makes the data essential to the researchers in Arabic natural language processing (NLP) due to the lack of resources in this language. The data is made publicly available to allow researchers to use it as a benchmark for …

Self-Optimizing Feature Generation Via Categorical Hashing Representation And Hierarchical Reinforcement Crossing, Wangyang Ying, Dongjie Wang, Kunpeng Liu, Leilei Sun, Yanjie Fu

Computer Science Faculty Publications and Presentations

Feature generation aims to generate new and meaningful features to create a discriminative representation space. A generated feature is meaningful when the generated feature is from a feature pair with inherent feature interaction. In the real world, experienced data scientists can identify potentially useful feature-feature interactions, and generate meaningful dimensions from an exponentially large search space in an optimal crossing form over an optimal generation path. But, machines have limited human-like abilities. We generalize such learning tasks as self-optimizing feature generation. Self-optimizing feature generation imposes several under-addressed challenges on existing systems: meaningful, robust, and efficient generation. To tackle these challenges, …

Geodatabase And Modeling Code Used For Dynamic Landslide Hazard Maps In Eastern Kentucky, Nathaniel O'Leary, L. Sebastian Bryson

Earth and Environmental Sciences Research Data

We developed spatiotemporal landslide hazard maps (LHMs) using soil, hydrologic, and geomorphic parameters from the subaerial infinite slope factor of safety (FS) equation under unsaturated conditions. Soil properties were extracted from the NRCS WSS, while geomorphic variables were derived from a 1.5 m LiDAR-based DEM and ArcGIS Online. Soil moisture from Hydrus-1D, driven by precipitation and evapotranspiration (ET) data from Irrigation Manager, introduced temporal variability. Validation against known landslide sites showed spatial and temporal FS accuracy despite some false positives. We also developed a landslide susceptibility maps (LSM) after comparing three machine learning algorithms, with bagged trees achieving the highest …

Enabling The Clinical Application Of Artificial Intelligence In Genomics: A Perspective Of The Amia Genomics And Translational Bioinformatics Workgroup, Nephi A Walton, Radha Nagarajan, Chen Wang, Murat Sincan, Robert R Freimuth, David B Everman, Derek C Walton, Scott P Mcgrath, Dominick J Lemas, Panayiotis V Benos, Alexander V Alekseyenko, Qianqian Song, Ece Gamsiz Uzun, Casey Overby Taylor, Alper Uzun, Thomas Nate Person, Nadav Rappoport, Zhongming Zhao, Marc S Williams

Faculty, Staff and Student Publications

OBJECTIVE: Given the importance AI in genomics and its potential impact on human health, the American Medical Informatics Association-Genomics and Translational Biomedical Informatics (GenTBI) Workgroup developed this assessment of factors that can further enable the clinical application of AI in this space.

PROCESS: A list of relevant factors was developed through GenTBI workgroup discussions in multiple in-person and online meetings, along with review of pertinent publications. This list was then summarized and reviewed to achieve consensus among the group members.

CONCLUSIONS: Substantial informatics research and development are needed to fully realize the clinical potential of such technologies. The development of …

Self-Consistent Phonon Calculations And Quartic Anharmonic Lattice Thermal Conductivity In 2d Ins Monolayer, Eesha Andharia, Hind Alqurashi, Abhiyan Pandit, Bothina Hamad

Physics Faculty Publications and Presentations

Low lattice thermal conductivity (κ_l) is a crucial factor for higher figure-of merit and hence the efficiency of thermoelectric generators. There are several reports on intrinsically low κ_l values in two-dimensional (2D) van der Waals materials using density functional theory and molecular dynamics simulations. In general, phonon dispersions are studied at absolute zero temperature using the finite-displacement approach within harmonic approximations. In addition, the κ_l is calculated using the third-order cubic interatomic force constants (IFCs) by solving the Boltzmann transport equation for phonons. In these calculations, we use quartic IFCs to solve self-consistent phonon equations …

Machine-Learning-Assisted Leak Detection Using Distributed Temperature And Acoustic Sensors, Hayden Gemeinhardt, Jyotsna Sharma

Faculty Publications

Small leaks, many of which often go undetected using conventional gauges, remain an urgent problem for an aging pipeline infrastructure. This article proposes a method that allows for an automated and robust leak detection and localization system by fusing distributed acoustic sensor (DAS) and distributed temperature sensor (DTS) data for machine learning. Distributed fiber-optic sensing creates an advantage over conventional gauges by providing real-time continuous measurements along the entire length of the fiber-instrumented pipeline at a high spatiotemporal resolution and sensitivity that can detect small leaks, as well as the leak location. High sensitivity, however, can create noisy data. Thus, …

Detection And Classification Of Sporadic E Using Convolutional Neural Networks, J. A. Ellis, Daniel J. Emmons, M. B. Cohen

Faculty Publications

In this work, convolutional neural networks (CNN) are developed to detect and characterize sporadic E (E_s), demonstrating an improvement over current methods. This includes a binary classification model to determine if Es is present, followed by a regression model to estimate the Es ordinary mode critical frequency (foEs), a proxy for the intensity, along with the height at which the Es layer occurs (hEs). Signal-to-noise ratio (SNR) and excess phase profiles from six Global Navigation Satellite System (GNSS) radio occultation (RO) missions during the years 2008–2022 are used as the inputs of the model. Intensity (foEs) and the …

Training Data Augmentation With Data Distilled By Principal Component Analysis, Nikolay Metodiev Sirakov, Tahsin Shahnewaz, Arie Nakhmmani

Faculty Publications

This work develops a new method for vector data augmentation. The proposed method applies principal component analysis (PCA), determines the eigenvectors of a set of training vectors for a machine learning (ML) method and uses them to generate the distilled vectors. The training and PCA-distilled vectors have the same dimension. The user chooses the number of vectors to be distilled and augmented to the set of training vectors. A statistical approach determines the lowest number of vectors to be distilled such that when augmented to the original vectors, the extended set trains an ML classifier to achieve a required accuracy. …

Emotion Recognition Using Various Measures And Computational Methods: Review, Abir Hamrouni, Fatima Bendella

Iraqi Journal for Computer Science and Mathematics

Emotion recognition has garnered significant attention as a burgeoning research domain, owing to its potential applications across diverse fields such ashuman-computer interaction, affective gaming, marketing, and human-robotinteraction.Accurately interpreting and appropriately responding to human emotions remains a critical challenge in the development of systems. This obstacle necessitates a thoroughunderstanding of emotions to enhance user experiences within suchsystems.This paper conducts a comprehensive review focusing on advancements in emotion recognition techniques, with an emphasis on leveraging a variety of sensors and computational methods. Our study findingshighlight the significantimprovement to emotion recognition accuracy when multiple measures and computational methods, rather than a single modality, …

A Generating Distorted Captcha Images Using A Machine Learning Algorithm, Saba Abdulbaqi Salman, Yasmin Makki Mohialden, Nadia Mahmood Hussien

Iraqi Journal for Computer Science and Mathematics

CAPTCHAs (Completely Automated Public Turing Test to Tell Computers and Humans Apart) have become universal in web security systems to differentiate between automated bots and human users. This research presents a novel approach for generating and classifying distorted CAPTCHA images utilizing machine learning techniques. The process involvesdeveloping a random text and rendering it onto an image, introducing distortion for security. The proposed method involves developing CAPTCHA images by combining text rendering and controlled distortion techniques. These images are then utilized to train a random forest classifier for accurate recognition. A Random Forest classifier is employed to recognize the generated CAPTCHA …

Real-Timelie-Speech Determination Using Voice-Stress Technology, Fadi K. Al-Dhaher, Duraid Y. Mohammed, Mohammed Khalaf, Khamis A. Al-Karawi, Mohammad Sarfraz, Muhammad Mazin Al Maathidi

Iraqi Journal for Computer Science and Mathematics

Lie detection has gained importance and is now extremely significant in a variety of fields. It plays an important role in several domains, including law enforcement, criminal investigations, national security,workplace ethics, and personal relationships. As advances in lie detection continue to develop, real-time approaches such asvoice stress technology have emerged as a feasible alternative to traditional methods such as polygraph testing. Polygraph testing, ahistorical and generally established approach, may be enhanced or replaced bythese revolutionary real-time techniques. Traditional lie detection procedures, such aspolygraph testing, have been challenged for their lack of reliability and validity. Newertechniques, such as brain imaging and …

Early Detection Of Cardiovascular Disease Utilizing Machine Learning Techniques: Evaluating The Predictive Capabilities Of Seven Algorithms, Maad M. Mijwil, Alaa K. Faieq, Mohammad Aljanabi

Iraqi Journal for Computer Science and Mathematics

Heart disease is the leading cause of death indeveloped countries, as it causes many deaths annually. Despite the availability of effective treatments, heartdisease remains a significant challenge to public health, so early detection is essential in enhancing patient outcomesand reducing mortality.Artificial intelligence seeks to help physicians make the right decisions about a patient's health condition. In this regard, the authors decided to utilize machine learning techniques (k-nearest neighbor, decision tree, linear regression, support vector machine, naïve bayes, multilayer perceptron, random forest) to contribute to the classification of the heart disease dataset, where it is determined whether a person is suffering …

Integrating Uav Hyperspectral Data And Radiative Transfer Model Simulation To Estimate Maize Leaf And Canopy Nitrogen Content, J. Li, Y Ge, L. A. Puntel, D. Heeren, G. Bai, G. R. Balboa, J. A. Gamon, T. J. Arkebauer, Y. Shi

School of Natural Resources: Faculty Publications

No abstract provided.

Piloting Of Specific Common Tools In Support Of Digital And Online Learning, Orlin Marinov, Karlis Valtins, Janina Fengel, K.C. O'Rourke, Lubomir Dimitrov, Rossen Radonov

Conference Proceedings

This publication is an insightful exploration into the evolving landscape of digital and online learning, with a specific focus on the pilot testing and implementation of innovative educational tools. It aims to critically assess the effectiveness of these tools in enhancing the educational experience and the overall learning outcomes in digital environments.

This manuscript is particularly relevant for educators, education technologists, policymakers, and scholars interested in the advancement of digital education. It serves as a pivotal resource, providing valuable insights into the utilization and effectiveness of digital tools in education. The publication's contribution to academic scholarship is significant, offering a …