Physical Sciences and Mathematics Commons^™

A Multistage Framework For Detection Of Very Small Objects, Duleep Rathgamage Don, Ramazan Aygun, Mahmut Karakaya

Published and Grey Literature from PhD Candidates

Small object detection is one of the most challenging problems in computer vision. Algorithms based on state-of-the-art object detection methods such as R-CNN, SSD, FPN, and YOLO fail to detect objects of very small sizes. In this study, we propose a novel method to detect very small objects, smaller than 8×8 pixels, that appear in a complex background. The proposed method is a multistage framework consisting of an unsupervised algorithm and three separately trained supervised algorithms. The unsupervised algorithm extracts ROIs from a high-resolution image. Then the ROIs are upsampled using SRGAN, and the enhanced ROIs are detected by our …

A New Kind Of Data Science: The Need For Ethical Analytics, Jonathan Boardman

Published and Grey Literature from PhD Candidates

Ethics can no longer be regarded as an add-on in data science and analytics. This paper argues for the necessity of formalizing a new, practically-oriented sub-discipline of AI ethics by outlining the needs, highlighting shortcomings in current approaches, and providing a framework for ethical analytics, which is concerned with the study of the ethical issues surrounding the development, deployment, and/or dissemination of ML/AI systems and data science research, as well as the development of tools and procedures to mitigate ethical harms. While data science and machine learning are primarily concerned with data from start to finish, ethical analytics is concerned …

Maker Math: Exploring Mathematics Through Digitally Fabricated Tools With K–12 In-Service Teachers, Jason R. Harron, Yi Jin, Amy F. Hillen, Lindsey Mason, Lauren Siegel

Faculty Open Access Publishing Fund Collection

This paper reports on nine elementary, middle, and high school in-service teachers who participated in a series of workshops aimed at exploring the wonder, joy, and beauty of mathematics through the creation and application of digitally fabricated tools (i.e., laser-cut and 3D printed). Using the Technological Pedagogical and Content Knowledge (TPACK) framework to investigate technological, pedagogical, contextual, and content knowledge, researchers applied qualitative methods to uncover the affordances and constraints of teaching and learning math concepts with digitally fabricated tools and examined how the workshops supported broadening participation in mathematics by focusing on the connections between mathematical inquiry, nature, and …

Qcd Corrections In Tqγ Production At Hadron Colliders, Nikolaos Kidonakis, Nodoka Yamanaka

Faculty and Research Publications

We study QCD corrections for the associated production of a single top quark and a photon (tqγ production) at hadron colliders. We calculate the NLO cross section at LHC and future collider energies for a variety of kinematical cuts, and we estimate uncertainties from scale dependence and from parton distributions. We also calculate differential distributions in top-quark transverse-momentum and rapidity as well as photon energy. Finally, we study higher-order corrections from soft-gluon emission for this process, and we provide approximate NNLO (aNNLO) results for the cross section and top-quark differential distributions. We also compare our calculations with recent measurements from …

Evaluation Of Chelating Agents Used In Phytoextraction By Switchgrass Of Lead Contaminated Soil, Genna Hart, Marina Koether, Thomas C. Mcelroy, Sigurdur Greipsson

Faculty Open Access Publishing Fund Collection

Soil lead (Pb) contamination is a recognized environmental and global health problem. Phytoextraction of Pb using switchgrass (Panicum virgatum L.), a second-generation biofuel crop, is typically enhanced by soil chelation. The effectiveness of four different chelating agents, phytic acid (inositol hexaphosphate), citric acid, NTA (nitrilotriacetic acid), and EDTA (ethylenediaminetetraacetic acid) was examined in pot culture. Plants treated with EDTA (1 mM) showed significantly higher shoot Pb concentrations compared to control plants and plants treated with other chelates. Lead-solubility following phytoextraction was examined by soil washing using 0.01 and 0.05 M acetic acid as an extractant solution revealed no …

Soil Chemistry And Clay Mineralogy Of An Alluvial Chronosequence From The North Carolina Sandhills Of The Upper Coastal Plain, Usa, Bradley E. Suther, David S. Leigh, Larry T. West

Faculty and Research Publications

Temporal changes in soil development were assessed on fluvial terraces of the Little River in the upper Coastal Plain of North Carolina. We examined five profiles from each of six surfaces spanning about 100,000 years. Soil-age relationships were evaluated with inter-surface clay mineral comparisons and regression of chemical properties versus previously reported optically-stimulated luminescence ages using the most developed subsoil horizon per profile. Bases to alumina (Bases/Al2 O3 ) ratios have negative correlations with age, whereas dithionite-Fe (FeD ) concentrations are positively correlated with time and differentiate floodplain (BP) from terrace (≥10 ± 2 ka) soils and T4 pedons (75 …

Directional Pairwise Class Confusion Bias And Its Mitigation, Sudhashree Sayenju, Ramazan Aygun Phd, Jonathan Boardman, Duleep Prasanna Rathgamage Don, Yifan Zhang Phd, Bill Franks, Sereres Johnston Phd, George Lee, Dan Sullivan, Girish Modgil Phd

Published and Grey Literature from PhD Candidates

Recent advances in Natural Language Processing have led to powerful and sophisticated models like BERT (Bidirectional Encoder Representations from Transformers) that have bias. These models are mostly trained on text corpora that deviate in important ways from the text encountered by a chatbot in a problem-specific context. While a lot of research in the past has focused on measuring and mitigating bias with respect to protected attributes (stereotyping like gender, race, ethnicity, etc.), there is lack of research in model bias with respect to classification labels. We investigate whether a classification model hugely favors one class with respect to another. …

Explainabilityaudit: An Automated Evaluation Of Local Explainability In Rooftop Image Classification, Duleep Rathgamage Don, Jonathan Boardman, Sudhashree Sayenju, Ramazan Aygun, Yifan Zhang, Bill Franks, Sereres Johnston, George Lee, Dan Sullivan, Girish Modgil

Published and Grey Literature from PhD Candidates

Explainable Artificial Intelligence (XAI) is a key concept in building trustworthy machine learning models. Local explainability methods seek to provide explanations for individual predictions. Usually, humans must check these explanations manually. When large numbers of predictions are being made, this approach does not scale. We address this deficiency for a rooftop classification problem specifically with ExplainabilityAudit, a method that automatically evaluates explanations generated by a local explainability toolkit and identifies rooftop images that require further auditing by a human expert. The proposed method utilizes explanations generated by the Local Interpretable Model-Agnostic Explanations (LIME) framework as the most important superpixels of …

Integrated Gradients Is A Nonlinear Generalization Of The Industry Standard Approach To Variable Attribution For Credit Risk Models, Jonathan Boardman, Md Shafiul Alam, Xiao Huang, Ying Xie

Published and Grey Literature from PhD Candidates

In modern society, epistemic uncertainty limits trust in financial relationships, necessitating transparency and accountability mechanisms for both consumers and lenders. One upshot is that credit risk assessments must be explainable to the consumer. In the United States regulatory milieu, this entails both the identification of key factors in a decision and the provision of consistent actions that would improve standing. The traditionally accepted approach to explainable credit risk modeling involves generating scores with Generalized Linear Models (GLMs) - usually logistic regression, calculating the contribution of each predictor to the total points lost from the theoretical maximum, and generating reason codes …

Kennesaw State University Hpc Facilities And Resources, Tom Boyle, Ramazan Aygun

Digital Commons Training Materials

The Kennesaw State University High Performance Computing (HPC) resources represent the University’s commitment to research computing. This resource contains verbiage for users of Kennesaw State University's HPC resources to include in their grants and publications. Please use the recommended citation rather than including the listed authors in the your citations.

The current version was published Fall 2023. Previous versions can be found below.

Bound States And Energy Eigenvalues Of A Radial Screened Coulomb Potential, Eric Stachura, N. Hancock

Faculty and Research Publications

We analyze bound states and other properties of solutions of a radial Schrödinger equation with a new screened Coulomb potential. In particular, we employ hypervirial relations to obtain eigen-energies for a Hydrogen atom with this potential. Additionally, we appeal to a sharp estimate for a modified Bessel function to estimate the ground state energy of such a system. Finally, when the angular quantum number ℓ ≠ 0, we obtain evidence for a critical screening parameter, above which bound states cease to exist.

Tz' Production At Hadron Colliders, Marco Guzzi, Nikolaos Kidonakis

Faculty and Research Publications

We study the production of a single top quark in association with a heavy extra Z' at hadron colliders in new physics models with and without flavor-changing neutral-current (FCNC) couplings. We use QCD soft-gluon resummation and threshold expansions to calculate higher-order corrections for the total cross section and transverse momentum distributions for t Z' production. The impact of the uncertainties due to the structure of the proton and scale dependence is also analyzed.

The Expanded View Of Individualism And Collectivism: One, Two, Or Four Dimensions?, Jennifer L. Priestley, Kamal Fatehi, Gita Taasoobshirazi

Faculty and Research Publications

Recent research to analyze and discuss cultural differences has employed a combination of five major dimensions of individualism–collectivism, power distance, uncertainty avoidance, femininity– masculinity (gender role differentiation), and long-term orientation. Among these dimensions, individualism–collectivism has received the most attention. Chronologically, this cultural attribute has been regarded as one, then two, and more recently, four dimensions of horizontal and vertical individualism and collectivism. However, research on this issue has not been conclusive and some have argued against this expansion. The current study attempts to explain and clarify this discussion by using a shortened version of the scale developed by Singelis et …

Ohpl Tutorial, Bob Vanderheyden, Ying Xie

OHPL Learning Materials

A tutorial for the OHPL Project

Developing And Improving Risk Models Using Machine-Learning Based Algorithms, Yan Wang, Sherry Ni

Published and Grey Literature from PhD Candidates

The objective of this study is to develop a good risk model for classifying business delinquency by simultaneously exploring several machine learning-based methods including regularization, hyperparameter optimization, and model ensembling algorithms. The rationale under the analyses is firstly to obtain good base binary classifiers (include Logistic Regression (LR), K-Nearest Neighbors (KNN ), Decision Tree (DT), and Artificial Neural Networks (ANN )) via regularization and appropriate settings of hyper-parameters. Then two model ensembling algorithms including bagging and boosting are performed on the good base classifiers for further model improvement. The models are evaluated using accuracy, Area Under the Receiver Operating Characteristic …

An Automatic Interaction Detection Hybrid Model For Bankcard Response Classification, Yan Wang, Sherry Ni, Brian Stone

Published and Grey Literature from PhD Candidates

Data mining techniques have numerous applications in bankcard response modeling. Logistic regression has been used as the standard modeling tool in the financial industry because of its almost always desirable performance and its interpretability. In this paper, we propose a hybrid bankcard response model, which integrates decision tree-based chi-square automatic interaction detection (CHAID) into logistic regression. In the first stage of the hybrid model, CHAID analysis is used to detect the possible potential variable interactions. Then in the second stage, these potential interactions are served as the additional input variables in logistic regression. The motivation of the proposed hybrid model …

A Two-Stage Hybrid Model By Using Artificial Neural Networks As Feature Construction Algorithms, Yan Wang, Sherry Ni, Brian Stone

Published and Grey Literature from PhD Candidates

We propose a two-stage hybrid approach with neural networks as the new feature construction algorithms for bankcard response classifications. The hybrid model uses a very simple neural network structure as the new feature construction tool in the first stage, then the newly created features are used as the additional input variables in logistic regression in the second stage. The model is compared with the traditional one-stage model in credit customer response classification. It is observed that the proposed two-stage model outperforms the one-stage model in terms of accuracy, the area under the ROC curve, and KS statistic. By creating new …

Predicting Class-Imbalanced Business Risk Using Resampling, Regularization, And Model Ensembling Algorithms, Yan Wang, Sherry Ni

Published and Grey Literature from PhD Candidates

We aim at developing and improving the imbalanced business risk modeling via jointly using proper evaluation criteria, resampling, cross-validation, classifier regularization, and ensembling techniques. Area Under the Receiver Operating Characteristic Curve (AUC of ROC) is used for model comparison based on 10-fold cross-validation. Two undersampling strategies including random undersampling (RUS) and cluster centroid undersampling (CCUS), as well as two oversampling methods including random oversampling (ROS) and Synthetic Minority Oversampling Technique (SMOTE), are applied. Three highly interpretable classifiers, including logistic regression without regularization (LR), L1-regularized LR (L1LR), and decision tree (DT) are implemented. Two ensembling techniques, including Bagging and Boosting, are …

A Xgboost Risk Model Via Feature Selection And Bayesian Hyper-Parameter Optimization, Yan Wang, Sherry Ni

Published and Grey Literature from PhD Candidates

This paper aims to explore models based on the extreme gradient boosting (XGBoost) approach for business risk classification. Feature selection (FS) algorithms and hyper-parameter optimizations are simultaneously considered during model training. The five most commonly used FS methods including weight by Gini, weight by Chi-square, hierarchical variable clustering, weight by correlation, and weight by information are applied to alleviate the effect of redundant features. Two hyper-parameter optimization approaches, random search (RS) and Bayesian tree-structuredParzen Estimator (TPE), are applied in XGBoost. The effect of different FS and hyper-parameter optimization methods on the model performance are investigated by the Wilcoxon Signed Rank …

Genetic Algorithm Guidance Of A Constraint Programming Solver For The Multiple Traveling Salesman Problem, Jessica M. Rudd, Andrew M. Henshaw, Lauren Staples, Sanjoosh Akkineni, Lin Li, Joe Demaio

Published and Grey Literature from PhD Candidates

This project developed a metaheuristic approach to the Multiple Traveling Salesman Problem that pairs a custom genetic algorithm with a conventional combinatorial optimization solver. This combined approach was used to build an optimal route for two popular radio show hosts to visit each of the 37 Atlanta area Jersey Mike's Subs in one day. This supported a fundraising eort to send children with chronic and terminal illnesses to Disney World through an organization called Bert's Big Adventure. Atlanta-area Jersey Mike's locations donated 100% of proceeds earned on this Day of Giving to Bert's Big Adventure. With the suggested route developed …

Fusion-Net: Integration Of Dimension Reduction And Deep Learning Neural Network For Image Classification, Mohammad Masum, Philippe Laval

Published and Grey Literature from PhD Candidates

Building a deep network using original digital images requires learning many parameters which may reduce the accuracy rates. The images can be compressed by using dimension reduction methods and extracted reduced features can be feeding into a deep network for classification. Hence, in the training phase of the network, the number of parameters will be decreased. Principal Component Analysis is a well-known dimension reduction technique that leverage orthogonal linear transformation of the original data. In this paper, we propose a neural network-based framework, named Fusion-Net, which implements PCA on an image dataset (CIFAR-10) and then a neural network applies on …

Evaluating The Impact Of Proactive Care Management With Idstrat, D.J. Donahue, Lauren Staples

Published and Grey Literature from PhD Candidates

This purpose of this study is to quantify potential cost savings and member care improvements as a result of engagement through BlueCross BlueShield of Tennessee’s (BCBST) Identification and Stratification (IDStrat) process. Commercial members engaged in clinical management that were identified through IDStrat were compared to commercial members identified through other means across several metrics including per-member, per-month (PMPM) cost and physician visits. Members identified by IDStrat experienced a statistically significant 7% greater reduction in costs after being engaged when compared with those identified by other methods. Members identified by IDStrat also experienced a significant reduction in emergency room visits after …

A Descriptive Study Of Variable Discretization And Cost-Sensitive Logistic Regression On Imbalanced Credit Data, Lili Zhang, Jennifer Priestley, Herman Ray, Soon Tan

Published and Grey Literature from PhD Candidates

Training classification models on imbalanced data tends to result in bias towards the majority class. In this paper, we demonstrate how variable discretization and cost-sensitive logistic regression help mitigate this bias on an imbalanced credit scoring dataset, and further show the application of the variable discretization technique on the data from other domains, demonstrating its potential as a generic technique for classifying imbalanced data beyond credit scoring. The performance measurements include ROC curves, Area under ROC Curve (AUC), Type I Error, Type II Error, accuracy, and F1 score. The results show that proper variable discretization and cost-sensitive logistic regression with …

The Sub-Eddington Boundary For The Quasar Mass–Luminosity Plane: A Theoretical Perspective, David Garofalo, Damian J. Christian, Andrew M. Jones

Faculty and Research Publications

By exploring more than sixty thousand quasars from the Sloan Digital Sky Survey Data Release 5, Steinhardt & Elvis discovered a sub-Eddington boundary and a redshift-dependent drop-off at higher black hole mass, possible clues to the growth history of massive black holes. Our contribution to this special issue of Universe amounts to an application of a model for black hole accretion and jet formation to these observations. For illustrativepurposes,we include~100,000 data points from the Sloan Digital Sky Survey Data Release 7 where the sub-Eddington boundary is also visible andpropose a theoretical picture that explains these features. By appealing to thin …

Self-Consistent Quantum-Kinetic Theory For Interplay Between Pulsed-Laser Excitation And Nonlinear Carrier Transport In A Quantum-Wire Array, Jeremy R. Gulley, Danghong Huang

Faculty and Research Publications

We propose a self-consistent many-body theory for coupling the ultrafast dipole-transition and carrier-plasma dynamics in a linear array of quantum wires with the scattering and absorption of ultrashort laser pulses. The quantum-wire non-thermal carrier occupations are further driven by an applied DC electric field along the wires in the presence of resistive forces from intrinsic phonon and Coulomb scattering of photo-excited carriers. The same strong DC field greatly modifies the non-equilibrium properties of the induced electron-hole plasma coupled to the propagating light pulse, while the induced longitudinal polarization fields of each wire significantly alters the nonlocal optical response from neighboring …

Search For X(3872) And X(3915) Decay Into Χc1Π0 In B Decays At Belle, The Belle Collaboration, Luminda Kulasiri

Faculty and Research Publications

We report a search for X(3872) and X (3915) in B+→χc1π0K+ decays. We set an upper limit of B(B+→X(3872)K+)×B(X(3872)→χc1π0)<8.1×10−6 and B(B+→X(3915)K+)×B(X(3915)→χc1π0)<3.8×10−5 at 90% confidence level. We also measure B(X(3872)→χc1π0)/B(X(3872)→J/ψπ+π−)<0.97 at 90% confidence level. The results reported here are obtained from 772×106 B¯B events collected at the Υ(4S) resonance with the Belle detector at the KEKB asymmetric-energy e+e− collider.

The Evolution Of Data Science: A New Mode Of Knowledge Production, Jennifer Lewis Priestley, Robert J. Mcgrath

Faculty and Research Publications

Is data science a new field of study or simply an extension or specialization of a discipline that already exists, such as statistics, computer science, or mathematics? This article explores the evolution of data science as a potentially new academic discipline, which has evolved as a function of new problem sets that established disciplines have been ill-prepared to address. The authors find that this newly-evolved discipline can be viewed through the lens of a new mode of knowledge production and is characterized by transdisciplinarity collaboration with the private sector and increased accountability. Lessons from this evolution can inform knowledge production …

Stochastic Ordering Of Pólya Random Variables And Monotonicity Of The Bernstein–Stancu Operator For A Negative Parameter, Florenţa Tripşa, Nicolae R. Pascu

Faculty and Research Publications

In the present paper, we prove that the probabilities of the Pólya urn distribution (with negative replacement) satisfy a monotonicity property similar to that of the binomial distribution. As a consequence, we show that the corresponding random variables are stochastically ordered with respect to the parameter giving the initial distribution of the urn. An equivalent formulation of this result shows that the new Bernstein–Stancu-type operator introduced in (Pascu et al. in Proc. Rom. Acad., Ser. A: Math. Phys. Tech. Sci. Inf. Sci. 2019, in press) is a monotone operator. The proofs are probabilistic in spirit and rely on various inequalities, …

Fr0 Radio Galaxies And Their Place In The Radio Morphology Classification, David Garofalo, Chandra B. Singh

Faculty and Research Publications

So-called FR0 radio galaxies have recently emerged as a family of active galaxies with all the same properties as FRI radio galaxies except for their ratio of core to total emission, which is about 30 times higher than that of FRI sources. We show how their properties fit within the gap paradigm as low, prograde, spinning black holes whose progenitors are powerful FRII quasars that transitioned rapidly from the cold mode into advection-dominated accretion over a few million years. The prediction is that if sufficient fuel exists, FR0 radio galaxies will evolve into full-fledged FRI radio galaxies and the observational …

Sustainability Education And Organizational Change: A Critical Case Study Of Barriers And Change Drivers At A Higher Education Institution, Edwin E. Akins Ii, Elizabeth Giddens, David Glassmeyer, Amy Gruss

Faculty and Research Publications

Integrating sustainability within institutions of higher education can have a tremendous impact on students, faculty, and the larger community. Sustainability efforts also experience many barriers to implementation within higher education contexts. A change management perspective can help characterize these barriers and ways to overcome them. In this critical case study, we use a process model to examine the kinds of barriers Kennesaw State University (KSU) has faced regarding implementation of academic sustainability and to evaluate change drivers that can advance sustainability during a time of leadership change. The process model evaluates barriers and change drivers according to published frameworks, and …