Statistics and Probability Commons^™

Direct Error Driven Learning For Deep Neural Networks With Applications To Bigdata, R. Krishnan, Jagannathan Sarangapani, V. A. Samaranayake

Electrical and Computer Engineering Faculty Research & Creative Works

In this paper, generalization error for traditional learning regimes-based classification is demonstrated to increase in the presence of bigdata challenges such as noise and heterogeneity. To reduce this error while mitigating vanishing gradients, a deep neural network (NN)-based framework with a direct error-driven learning scheme is proposed. To reduce the impact of heterogeneity, an overall cost comprised of the learning error and approximate generalization error is defined where two NNs are utilized to estimate the costs respectively. To mitigate the issue of vanishing gradients, a direct error-driven learning regime is proposed where the error is directly utilized for learning. It …

A Multi-Step Nonlinear Dimension-Reduction Approach With Applications To Bigdata, R. Krishnan, V. A. Samaranayake, Jagannathan Sarangapani

Mathematics and Statistics Faculty Research & Creative Works

In this paper, a multi-step dimension-reduction approach is proposed for addressing nonlinear relationships within attributes. In this work, the attributes in the data are first organized into groups. In each group, the dimensions are reduced via a parametric mapping that takes into account nonlinear relationships. Mapping parameters are estimated using a low rank singular value decomposition (SVD) of distance covariance. Subsequently, the attributes are reorganized into groups based on the magnitude of their respective singular values. The group-wise organization and the subsequent reduction process is performed for multiple steps until a singular value-based user-defined criterion is satisfied. Simulation analysis is …

A Study Of Flight Simulation Training Time, Aircraft Training Time, And Pilot Competence As Measured By The Naval Standard Score, Aaron D. Judy

Doctor of Education (Ed.D)

The purpose of the study was to investigate the relationships between US Navy T-45C flight simulation training time, actual aircraft training time, and intermediate and advanced jet pilot competence as measured by the Naval Standard Score (NSS). Examining the relationships between US Navy T-45C flight simulation time and actual aircraft flight time may provide further information on flight simulation training versus actual aircraft training to aviation authorities, flight instructors, the military aviation community, the commercial aviation community, and academia. The study was non-experimental, correlational, causal-comparative with an emphasis upon the establishment of mathematic and predictive relationships using archival data from …

They're Only Nuclear Weapons: An Exploratory Analysis Of Safety Climate Within The Nuclear Enterprise, Brandon M. Clements

Theses and Dissertations

By possessing nuclear weapons, the United States Air Force is inherently exposed to extreme safety concerns. With multiple setbacks in recent years (e.g., unauthorized transport of nuclear weapons, cheating scandals, and career dissatisfaction), some have begun to wonder how safe the nuclear enterprise truly is. Building upon the concept of safety climate, this study explores safety climate constructs and trends associated with current nuclear maintenance safety climate survey data.

Effect Of Anthropometric Variability On Middle-Market Aircraft Seating, Tara C. Sriram

International Journal of Aviation, Aeronautics, and Aerospace

A middle-of-market aircraft, or MoMA, is defined as an aircraft capable of flying 180-250 passengers without refueling for 2,300-5,800 miles(~2,000-5,000 nautical miles). As the name suggests, middle-of-market aircraft are positioned in between the market segments served by narrow body (single-aisle) and wide body (twin-aisle) aircraft. This paper presents the findings of a study on the effect of anthropomorphic variability on economy class seating on middle-market aircraft currently in service. The study found that among 130 middle-market LOPAs, the mean seat pitch was greater for US airlines than for Asian airlines. Furthermore, the sampled Asian airlines had a higher preference …

Monte Carlo Simulations Of Three-Dimensional Electromagnetic Gaussian Schell-Model Sources, Milo W. Hyde Iv, Santasri Bose-Pillai, Olga Korotkova

Faculty Publications

This article presents a method to simulate a three-dimensional (3D) electromagnetic Gaussian-Schell model (EGSM) source with desired characteristics. Using the complex screen method, originally developed for the synthesis of two-dimensional stochastic electromagnetic fields, a set of equations is derived which relate the desired 3D source characteristics to those of the statistics of the random complex screen. From these equations and the 3D EGSM source realizability conditions, a single criterion is derived, which when satisfied guarantees both the realizability and simulatability of the desired 3D EGSM source. Lastly, a 3D EGSM source, with specified properties, is simulated; the Monte Carlo simulation …

Phytoforensics: Trees As Bioindicators Of Potential Indoor Exposure Via Vapor Intrusion, Jordan L. Wilson, V. A. Samaranayake, Matt A. Limmer, Joel Gerard Burken

Mathematics and Statistics Faculty Research & Creative Works

Human exposure to volatile organic compounds (VOCs) via vapor intrusion (VI) is an emerging public health concern with notable detrimental impacts on public health. Phytoforensics, plant sampling to semi-quantitatively delineate subsurface contamination, provides a potential non-invasive screening approach to detect VI potential, and plant sampling is effective and also time- and cost-efficient. Existing VI assessment methods are time- and resource-intensive, invasive, and require access into residential and commercial buildings to drill holes through basement slabs to install sampling ports or require substantial equipment to install groundwater or soil vapor sampling outside the home. Tree-core samples collected in 2 days at …

A Preliminary Study Of Smithport Plain Bottle Morphology In The Southern Caddo Area, Robert Z. Selden Jr.

CRHR: Archaeology

This study expands upon a previous analysis of the Clarence H. Webb collection, which resulted in the identification of two discrete shapes used in the manufacture of the base and body of Smithport Plain bottles. The sample includes the Smithport Plain bottles from the Webb collection, and four new bottles: two previously repatriated specimens in the Pohler Collection, and two from the Mitchell site (41BW4) to test whether those specimens align morphologically with the Belcher Mound or Smithport Landing specimens. Results indicate significant allometry and a significant difference in Smithport Plain body and base shapes for bottles produced at the …

A Proposed Taxonomy For The Systems Statistical Engineering Body Of Knowledge, Teddy Steven Cotter

Engineering Management & Systems Engineering Faculty Publications

In the ASEM-IAC 2012, Cotter (2012) identified the gaps in knowledge that statistical engineering needs to address, explored additional gaps in knowledge not addressed in the prior works, and set forth a working definition of and body of knowledge for statistical engineering. In the ASEM-IAC 2015, Cotter (2015) proposed a systemic causal Bayesian hierarchical model that addressed the knowledge gap needed to integrate deterministic mathematical engineering causal models within a stochastic framework. Missing, however, is the framework for specifying the hierarchical qualitative systems structures necessary and sufficient for specifying systemic causal Bayesian hierarchical models. In the ASEM-IAC 2016, Cotter (2016) …

Mixture Models With Grouping Structure: Retail Analytics Applications, Haidar Almohri

Wayne State University Dissertations

Growing competitiveness and increasing availability of data is generating tremendous interest in data-driven analytics across industries. In the retail sector, stores need targeted guidance to improve both the efficiency and effectiveness of individual stores based on their specific location, demographics, and environment. We propose an effective data-driven framework for internal benchmarking that can lead to targeted guidance for individual stores. In particular, we propose an objective method for segmenting stores using a model-based clustering technique that accounts for similarity in store performance dynamics. It relies on effective Finite Mixture of Regression (FMR) techniques for carrying out the model-based clustering with …

Effect Of Socioeconomic And Demographic Factors On Kentucky Crashes, Aaron Berry Cambron

Theses and Dissertations--Civil Engineering

The goal of this research was to examine the potential predictive ability of socioeconomic and demographic data for drivers on Kentucky crash occurrence. Identifying unique background characteristics of at-fault drivers that contribute to crash rates and crash severity may lead to improved and more specific interventions to reduce the negative impacts of motor vehicle crashes. The driver-residence zip code was used as a spatial unit to connect five years of Kentucky crash data with socioeconomic factors from the U.S. Census, such as income, employment, education, age, and others, along with terrain and vehicle age. At-fault driver crash counts, normalized over …

Location Optimization Of A Coal Power Plant To Balance Coal Supply And Electric Transmission Costs Against Plant’S Emission Exposure, Najam Khan

Electronic Theses and Dissertations

This research is focused on developing a location analysis methodology that can minimize the pollutant exposure to the public while ensuring that the combined costs of electric transmission losses and coal logistics are minimized. Coal power plants will provide a critical contribution towards meeting electricity demands for various nations in the foreseeable future. The site selection for a new coal power plant is extremely important from an investment point of view. The operational costs for running a coal power plant can be minimized by a combined emphasis on placing a coal power plant near coal mines as well as customers. …

Automated Tree-Level Forest Quantification Using Airborne Lidar, Hamid Hamraz

Theses and Dissertations--Computer Science

Traditional forest management relies on a small field sample and interpretation of aerial photography that not only are costly to execute but also yield inaccurate estimates of the entire forest in question. Airborne light detection and ranging (LiDAR) is a remote sensing technology that records point clouds representing the 3D structure of a forest canopy and the terrain underneath. We present a method for segmenting individual trees from the LiDAR point clouds without making prior assumptions about tree crown shapes and sizes. We then present a method that vertically stratifies the point cloud to an overstory and multiple understory tree …

Analyzing Sensor Based Human Activity Data Using Time Series Segmentation To Determine Sleep Duration, Yogesh Deepak Lad

Masters Theses

"Sleep is the most important thing to rest our brain and body. A lack of sleep has adverse effects on overall personal health and may lead to a variety of health disorders. According to Data from the Center for disease control and prevention in the United States of America, there is a formidable increase in the number of people suffering from sleep disorders like insomnia, sleep apnea, hypersomnia and many more. Sleep disorders can be avoided by assessing an individual's activity over a period of time to determine the sleep pattern and duration. The sleep pattern and duration can be …

A Model To Predict Concentrations And Uncertainty For Mercury Species In Lakes, Ashley Hendricks

Dissertations, Master's Theses and Master's Reports

To increase understanding of mercury cycling, a seasonal mass balance model was developed to predict mercury concentrations in lakes and fish. Results indicate that seasonality in mercury cycling is significant and is important for a northern latitude lake. Models, when validated, have the potential to be used as an alternative to measurements; models are relatively inexpensive and are not as time intensive. Previously published mercury models have neglected to perform a thorough validation. Model validation allows for regulators to be able to make more informed, confident decisions when using models in water quality management. It is critical to quantify uncertainty; …

An Efficient Method For Online Identification Of Steady State For Multivariate System, Honglun None Xu

Open Access Theses & Dissertations

Most of the existing steady state detection approaches are designed for univariate signals. For multivariate signals, the univariate approach is often applied to each process variable and the system is claimed to be steady once all signals are steady, which is computationally inefficient and also not accurate. The article proposes an efficient online method for multivariate steady state detection. It estimates the covariance matrices using two different approaches, namely, the mean-squared-deviation and mean-squared-successive-difference. To avoid the usage of a moving window, the process means and the two covariance matrices are calculated recursively through exponentially weighted moving average. A likelihood ratio …

Extraction Of Fiber Morphology From Sem Images For Quality Control Of Fiber Reinforced Composites Manufacturing, Md Fashiar Rahman

Open Access Theses & Dissertations

The morphology of fibers (e.g. spatial uniformity, orientation, and length) plays a decisive role in determining the material properties or fabrication quality of fiber-reinforced nanocomposites. Hence, determining the morphology becomes a very critical issue in the field of nanocomposite quality control. The conventional way of quality inspection is to take the scanning electron microscopic (SEM) images of the cross-section of composite material and do the visual checking of these SEM images to evaluate the nanofiber alignment and length distribution. But this type of inspection is often subjective, inaccurate and time consuming. Moreover, the extremely small size of nanofibers makes the …

Impact Of Highway Work Zones On Traffic Crashes: A Case Study In Michigan, Qadri Hafez Shaheen

Master's Theses and Doctoral Dissertations

Infrastructure in the US is severely aged and outdated. This presents a seemingly paradoxical problem in the field of construction management: In order to fix and make roads and highways more safe, construction zones must become inherently less safe in the process. There is a high cost to taxpayers and drivers, as work zones experience a significant amount of crashes and fatalities each year. To mitigate some of the factors that contribute to these crashes, this paper attempts to deliver guidelines on how to update relevant crash data, identify relevant factors, and create recommendations accordingly. The research focused particularly on …

Adaptive Methods For Point Cloud And Mesh Processing, Zinat Afrose

Computational Modeling & Simulation Engineering Theses & Dissertations

Point clouds and 3D meshes are widely used in numerous applications ranging from games to virtual reality to autonomous vehicles. This dissertation proposes several approaches for noise removal and calibration of noisy point cloud data and 3D mesh sharpening methods. Order statistic filters have been proven to be very successful in image processing and other domains as well. Different variations of order statistics filters originally proposed for image processing are extended to point cloud filtering in this dissertation. A brand-new adaptive vector median is proposed in this dissertation for removing noise and outliers from noisy point cloud data.

The major …

Particle Filters For State Estimation Of Confined Aquifers, Graeme Field

UNF Graduate Theses and Dissertations

Mathematical models are used in engineering and the sciences to estimate properties of systems of interest, increasing our understanding of the surrounding world and driving technological innovation. Unfortunately, as the systems of interest grow in complexity, so to do the models necessary to accurately describe them. Analytic solutions for problems with such models are provably intractable, motivating the use of approximate yet still accurate estimation techniques. Particle filtering methods have emerged as a popular tool in the presence of such models, spreading from its origins in signal processing to a diverse set of fields throughout engineering and the sciences including …

Unmasking Cost Growth Behavior: A Longitudinal Study, Cory N. D'Amico, Edward D. White, Jonathan D. Ritschel, Scott R. Kozlak

Faculty Publications

This article examines how cost growth factors (CGF) change over a program’s acquisition life cycle for 36 Department of Defense aircraft programs. Starting from Milestone B, the authors examine CGFs at five gateways: Critical Design Review, First Flight (FF), the end of Developmental Test and Evaluation (DT&E), Initial Operational Capability, and Full Operational Capability. Each CGF is assigned a color rating based upon the program’s cost growth: Green (low), Amber (moderate), or Red (high). Significant findings include dependencies among similar CGF color ratings and cost growth occurring primarily between FF and the end of DT&E during a program’s life cycle.

Offline And Online Density Estimation For Large High-Dimensional Data, Aref Majdara

Dissertations, Master's Theses and Master's Reports

Density estimation has wide applications in machine learning and data analysis techniques including clustering, classification, multimodality analysis, bump hunting and anomaly detection. In high-dimensional space, sparsity of data in local neighborhood makes many of parametric and nonparametric density estimation methods mostly inefficient.

This work presents development of computationally efficient algorithms for high-dimensional density estimation, based on Bayesian sequential partitioning (BSP). Copula transform is used to separate the estimation of marginal and joint densities, with the purpose of reducing the computational complexity and estimation error. Using this separation, a parallel implementation of the density estimation algorithm on a 4-core CPU is …

Application Of Remote Sensing And Machine Learning Modeling To Post-Wildfire Debris Flow Risks, Priscilla Addison

Dissertations, Master's Theses and Master's Reports

Historically, post-fire debris flows (DFs) have been mostly more deadly than the fires that preceded them. Fires can transform a location that had no history of DFs to one that is primed for it. Studies have found that the higher the severity of the fire, the higher the probability of DF occurrence. Due to high fatalities associated with these events, several statistical models have been developed for use as emergency decision support tools. These previous models used linear modeling approaches that produced subpar results. Our study therefore investigated the application of nonlinear machine learning modeling as an alternative. Existing models …

Wildfire Emissions In The Context Of Global Change And The Implications For Mercury Pollution, Aditya Kumar

Dissertations, Master's Theses and Master's Reports

Wildfires are episodic disturbances that exert a significant influence on the Earth system. They emit substantial amounts of atmospheric pollutants, which can impact atmospheric chemistry/composition and the Earth’s climate at the global and regional scales. This work presents a collection of studies aimed at better estimating wildfire emissions of atmospheric pollutants, quantifying their impacts on remote ecosystems and determining the implications of 2000s-2050s global environmental change (land use/land cover, climate) for wildfire emissions following the Intergovernmental Panel on Climate Change (IPCC) A1B socioeconomic scenario.

A global fire emissions model is developed to compile global wildfire emission inventories for major atmospheric …

Load Model Verification, Validation And Calibration Framework By Statistical Analysis On Field Data, Xiangqing Jiao, Yuan Liao, Thai Nguyen

Electrical and Computer Engineering Faculty Publications

Accurate load models are critical for power system analysis and operation. A large amount of research work has been done on load modeling. Most of the existing research focuses on developing load models, while little has been done on developing formal load model verification and validation (V&V) methodologies or procedures. Most of the existing load model validation is based on qualitative rather than quantitative analysis. In addition, not all aspects of model V&V problem have been addressed by the existing approaches. To complement the existing methods, this paper proposes a novel load model verification and validation framework that can systematically …

Improving Service Level Of Free-Floating Bike Sharing Systems, Aritra Pal

USF Tampa Graduate Theses and Dissertations

Bike Sharing is a sustainable mode of urban mobility, not only for regular commuters but also for casual users and tourists. Free-floating bike sharing (FFBS) is an innovative bike sharing model, which saves on start-up cost, prevents bike theft, and offers significant opportunities for smart management by tracking bikes in real-time with built-in GPS. Efficient management of a FFBS requires: 1) analyzing its mobility patterns and spatio-temporal imbalance of supply and demand of bikes, 2) developing strategies to mitigate such imbalances, and 3) understanding the causes of a bike getting damaged and developing strategies to minimize them. All of these …

An Enhanced Bridge Weigh-In-Motion Methodology And A Bayesian Framework For Predicting Extreme Traffic Load Effects Of Bridges, Yang Yu

LSU Doctoral Dissertations

In the past few decades, the rapid growth of traffic volume and weight, and the aging of transportation infrastructures have raised serious concerns over transportation safety. Under these circumstances, vehicle overweight enforcement and bridge condition assessment through structural health monitoring (SHM) have become critical to the protection of the safety of the public and transportation infrastructures. The main objectives of this dissertation are to: (1) develop an enhanced bridge weigh-in-motion (BWIM) methodology that can be integrated into the SHM system for overweight enforcement and monitoring traffic loading; (2) present a Bayesian framework to predict the extreme traffic load effects (LEs) …

Open Source Artificial Intelligence In A Biological/Ecological Context, Trevor Grant

Annual Symposium on Biomathematics and Ecology Education and Research

No abstract provided.

Deep Learning In R, Troy Hernandez

Annual Symposium on Biomathematics and Ecology Education and Research

No abstract provided.

Rates And Causes Of Accidents For General Aviation Aircraft Operating In A Mountainous And High Elevation Terrain Environment, Marisa Aguiar, Alan Stolzer, Douglas D. Boyd

Publications

Flying over mountainous and/or high elevation terrain is challenging due to rapidly changeable visibility, gusty/rotor winds and downdrafts and the necessity of terrain avoidance. Herein, general aviation accident rates and mishap cause/factors were determined (2001–2014) for a geographical region characterized by such terrain.

Accidents in single piston engine-powered aircraft for states west of the US continental divide characterized by mountainous terrain and/or high elevation (MEHET) were identified from the NTSB database. MEHET-related-mishaps were defined as satisfying any one, or more, criteria (controlled flight into terrain/obstacles (CFIT), downdrafts, mountain obscuration, wind-shear, gusting winds, whiteout, instrument meteorological conditions; density altitude, dust-devil) cited …