phd data science kennesaw state

Home > CCSE > Data Science and Analytics > Published and Grey Literature

Published and Grey Literature from PhD Candidates

The PhD Website

The Ph.D. in Analytics and Data Science is an advanced degree with a dual focus of application and research - where students will engage in real world business problems, which will inform and guide their research interests.

To ensure that our Ph.D. students in Analytics and Data Science are exposed to the latest issues and challenges of working across a wide variety of data contexts, individuals will be required to engage with one (or more) of the dozens of organizations which have agreed to sponsor doctorate-level projects for a minimum of three semesters (9 credit hours of engagement + 15 credit hours of dissertation research). These organizations span the continuum of application domains, including health care, banking, retail, government, and consumer finance. Students will also continue to work with the faculty adviser through their final year of project engagement and dissertation research.

The materials in this collection consists of research conducted by PhD candidates as a means to showcase the important work being done in the program.

Submissions from 2023 2023

A Multistage Framework for Detection of Very Small Objects , Duleep Rathgamage Don, Ramazan Aygun, and Mahmut Karakaya

Submissions from 2022 2022

Applications of Integrated Gradients in Credit Risk Modeling , Md Shafiul Alam, Jonathan Boardman, Xiao Huang, and Matthew Turner

A New Kind of Data Science: The Need for Ethical Analytics , Jonathan Boardman

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, and Ying Xie

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, and Girish Modgil

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, and Girish Modgil PhD

Submissions from 2020 2020

Fusion-Net: Integration of Dimension Reduction and Deep Learning Neural Network for Image Classification , Mohammad Masum and Philippe Laval

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, and Joe DeMaio

A XGBoost risk model via feature selection and Bayesian hyper-parameter optimization , Yan Wang and Sherry Ni

Developing and improving risk models using machine-learning based algorithms , Yan Wang and Sherry Ni

Predicting class-imbalanced business risk using resampling, regularization, and model ensembling algorithms , Yan Wang and Sherry Ni

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

A two-stage hybrid model by using artificial neural networks as feature construction algorithms , Yan Wang, Sherry Ni, and Brian Stone

Improving Investment Suggestions for Peer-to-Peer (P2P) Lending via Integrating Credit Scoring into Profit Scoring , Yan Wang and Xuelei Sherry Ni

Risk Prediction of Peer-to-Peer Lending Market by a LSTM Model with Macroeconomic Factor , Yan Wang and Xuelei Sherry Ni

Improving risk modeling via feature selection, hyper-parameter adjusting, and model ensembling , Yan Wang, Xuelei Sherry Ni, and Jennifer Priestley

Submissions from 2019 2019

Radically Simplifying Gated Recurrent Architectures Without Loss of Performance , Jonathan Boardman and Ying Xie

Evaluating the Impact of Proactive Care Management with IDStrat , D.J. Donahue and Lauren Staples

Outcome Prediction in Intensive Care Unit Settings with Claims Data , Lauren Staples and Ryan Rimby

A Product Affinity Segmentation Framework , Lili Zhang, Jennifer Priestley, Joseph DeMaio, and Sherry Ni

A Descriptive Study of Variable Discretization and Cost-Sensitive Logistic Regression on Imbalanced Credit Data , Lili Zhang, Jennifer Priestley, Herman Ray, and Soon Tan

Submissions from 2018 2018

A Comparison of Machine Learning Algorithms for Prediction of Past Due Service in Commercial Credit , Liyuan Liu M.A, M.S. and Jennifer Lewis Priestley Ph.D.

Automatic Knowledge Extraction from OCR Documents Using Hierarchical Document Analysis , Mohammad Masum, Sai Kosaraju, Tanju Bayramoglu, Girish Modgil, and Mingon Kang

The Validity of Online Patient Ratings of Physicians: Analysis of Physician Peer Reviews and Patient Ratings , Jennifer L. Priestley, Yiyun Zhou, and Robert McGrath

A Comparison of the Predictive Ability of Logistic Regression and Time Series Analysis on Business Credit Data , Lauren Staples

COMPARISON OF BANKRUPTCY PREDICTION MODELS WITH PUBLIC RECORDS AND FIRMOGRAPHICS , Lili Zhang, Jennifer Priestley, and Xuelei Ni

Influence of the Event Rate on Discrimination Abilities of Bankruptcy Prediction Models , Lili Zhang, Jennifer Priestley, and Xuelei Ni

Submissions from 2017 2017

Application of Support Vector Machine Modeling and Graph Theory Metrics for Disease Classification , Jessica M. Rudd

A Comparison of Decision Tree with Logistic Regression Model for Prediction of Worst Non-Financial Payment Status in Commercial Credit , Jessica M. Rudd MPH, GStat and Jennifer L. Priestley

Logistic Ensemble Models , Bob Vanderheyden and Jennifer L. Priestley

Binary Classification on Past Due of Service Accounts using Logistic Regression and Decision Tree , Yan Wang and Jennifer L. Priestley

A Sentiment-Change-Driven Event Discovery System , Lili Zhang, Ying Xie, and Guoliang Liu

Submissions from 2016 2016

An Analysis of Accuracy using Logistic Regression and Time Series , Edwin Baidoo and Jennifer L. Priestley

A Comparison of Machine Learning Techniques and Logistic Regression Method for the Prediction of Past-Due Amount , Jie Hao and Jennifer L. Priestley

Application of Isotonic Regression in Predicting Business Risk Scores , Linh T. Le and Jennifer L. Priestley

Advanced Search

• Notify me via email or RSS
• All Collections
• Disciplines
• Conferences
• Faculty Works
• Open Access
• Research Support
• Student Works
• Data Science Homepage

Useful Links

• Training Materials

Home | About | FAQ | My Account | Accessibility Statement

Privacy Copyright DigitalCommons@Kennesaw State University ISSN: 2576-6805

Faculty Web

• Current Students
• Online Only Students
• Faculty & Staff
• Parents & Family
• Alumni & Friends
• Community & Business
• Student Life
• Resources and Links
• Faculty Web Pages

Sherrill Hayes

Phone: 470-578-6499

Email: [email protected]

Office Location: KH 4428

Office Hours: By appointment

Dr. Sherrill W. Hayes serves as Interim Assistant Vice Provost of Student Success and Professor of Conflict Management at Kennesaw State University. He has held several leadership positions at Kennesaw State including founding Director of the School of Data Science and Analytics  (2021-2023), Director of the PhD in Analytics and Data Science (2018-2020), Director of the Master of Science in Conflict Management ( 2012-2017) . He was also Assistant Professor of Conflict Resolution at the University of North Carolina at Greensboro (2006-2012) , Visiting Assistant Professor of Human Development and Family Studies at the University of North Carolina at Greensboro (2003-2006), Lecturer of Early Childhood Education at Guilford Technical Community College (2004-2005), and has had part-time and visiting faculty positions at Gateshead College (UK), Newcastle University (UK), and Saarland University (Germany). He has taught over 50 different courses and supervised dozens of masters’ and undergraduate research projects and Ph.D. dissertations. 

His scholarship has resulted in over 25 publications, more than 50 conference presentations, and invitations to work with organizations and educational institutions in places such as Germany, Ghana, and Haiti. Dr. Hayes also has extensive experience as a practitioner of conflict management. He has worked as a family mediator, parenting coordinator, family business consultant, arbitrator, community mediator, group faciliator, and trainer. 

He is co-editor of the 2018 book - "Atone: Religion, Conflict, and Reconciliation" and is Associate Editor of the Journal of Peacebuilding and Development.  Dr. Hayes received his BS ('97) and MS ('00) in Human Development and Family Studies from the University of North Carolina at Greensboro and his Ph.D. in Sociology & Social Policy  from Newcastle University, UK ('05).

Dr. Hayes's CV .

Contact Info

Kennesaw Campus 1000 Chastain Road Kennesaw, GA 30144

Marietta Campus 1100 South Marietta Pkwy Marietta, GA 30060

Campus Maps

Phone 470-KSU-INFO (470-578-4636)

kennesaw.edu/info

Media Resources

Resources For

Related Links

• Financial Aid
• Degrees, Majors & Programs
• Job Opportunities
• Campus Security
• Global Education
• Sustainability
• Accessibility

470-KSU-INFO (470-578-4636)

© 2024 Kennesaw State University. All Rights Reserved.

• Privacy Statement
• Accreditation
• Emergency Information
• Report a Concern
• Open Records
• Human Trafficking Notice

• Data Science and Analytics
• All courses
• Online Certificate in Applied Data Science using Python

• Offerings & Schedules
• Financial Aid

The Online Certificate in Data Science using Python has been developed for working professionals who have some previous exposure to basic statistics. Students are encouraged to have completed at least one course in College Algebra or higher.

Required Courses:

• Introduction to Python Course
• Data Analysis and Probability
• Statistical Methods
• Advanced Methods

No application is required to register for this course.

Hardware Requirements : You must have a computer and some kind of high-speed internet connection – we recommend a 128mbps connection at a minimum.

Software Requirements : Please note that all work is completed using Python. This is an open source software, so it is free.

If you are a current KSU student in one of these four courses (Master in Applied Statistics, Master in Computer Science, PhD in Analytics and Data Science, PhD in Computer Science), please contact Cara Reeve ([email protected]) to enroll.

For companies requiring invoicing, please call 470-578-6765.

Data Science Online

#1: Online Certificate in Applied Data Science using Python

Financial aid options.

• Education Loans
• Click here for more information about Financial Aid

More Info / Contact Us

If you would like to receive more information about this program or speak with the program lead, submit your information below:

• More Details

Related Courses

Online certificate in applied data science using rapidminer, sql introduction – online, python developer certificate – online, javascript introduction – online, java introduction – online, online certificate in applied statistics using r.

Program Description 

Kennesaw State University’s Ph.D. with a major in Data Science and Analytics is an advanced degree, which trains individuals to translate large, structured and unstructured, complex data into information to improve decision-making, and become independent researchers. This highly interdisciplinary curriculum includes heavy emphasis on programming, machine learning, artificial intelligence, data mining, statistical modeling, and the mathematical foundations to support these concepts. The program also emphasizes communication skills, data ethics, and application of results to business and research problems. Graduates can pursue a position in the private or public sector as a “practicing” Data Scientist or a position within academia, where they are uniquely qualified to teach the next generation of data scientists.

return to top ↑

Admission, Enrollment, and Graduation Policies

Admission requirements.

The following are requirements beyond the general Admissions    requirements  

• GRE Score Report - Minimum Quantitative score is 160. Preferred Analytical Writing Score of 3.5.
• Resume or CV
• Statement of Intent describing how this degree facilitates your career goals. 
• At least one must be from an academic source.
• At least one must be from a source outside of the academic community.  
• Successful completion of Math courses through Calculus II 
• Proficiency in at least one analytical programming language (e.g., Python, SAS, R).

Admission Criteria for Unique Cases

Currently, there are no exceptions to the admission requirements.

Transfer Credit

No credit from outside institutions is accepted for this degree program.

Graduation Requirements

Each candidate must petition to graduate online. For more information, please view the corresponding section of Academic Policies:  5.0 PROGRAM REQUIREMENTS & GRADUATION   . 

Program Course Requirements

Required core (24 credits).

• CS 8265:Advanced Big Data Analytics
• CS 8267:Advanced Machine Learning
• MATH 8020:Graph Theory
• MATH 8030:Applied Discrete & Combinatorial Mathematics for Data Analysts
• STAT 8240:Data Mining I
• STAT 8250:Data Mining II
• DS 9700:Doctoral Internship (repeat for a total of 6 credits)

Electives and Concentration (21 hours)

Students can take up to 9 credit hours for 6000 or 7000 level courses in DS, STAT, or CS with permission of the program director. Students can take any 8000 or 9000 level course in DS, STAT, MATH, CS, or IT (other disciplines by permission of the director).

Computer Science Concentration

Students interested in pursuing a concentration in Computer Science must take at least 15 credit hours in CS courses at 8000 or 9000 levels (except CS 9900).

Statistics Concentration

Students interested in pursuing a concentration in Statistics must take at least 15 credit hours in STAT courses at 8000 or 9000 levels.

Research (33 hours)

• DS 9700:Doctoral Internship
• DS 9900:PhD Dissertation Research

Program Total (78 Credit Hours)

• Current Students
• Online Only Students
• Faculty & Staff
• Parents & Family
• Alumni & Friends
• Community & Business
• Student Life
• College of Computing and Software Engineering
• Executive Advisory Board
• CCSE Job Openings
• Academic Advising
• Student Resources
• International Student Resources
• Faculty Resources
• School of Data Science and Analytics
• Department of Computer Science
• Department of Information Technology
• Department of Software Engineering and Game Development
• Undergraduate
• Why Partner?
• Ways to Engage
• Friends & Corporate Affiliates
• K-12 outreach
• Employer Networking
• Data Science and Analytics
• Creating and Delivering an Effective Data-Driven Presentation

Same AI + Different Deployment Plans = Different Ethics

KENNESAW, Ga. | May 14, 2024

Why Autonomous Cars Aren’t Yet Ethical For Wide Deployment

This month I will address an aspect of the ethics of artificial intelligence (AI) and analytics that I think many people don’t fully appreciate. Namely, the ethics of a given algorithm can vary based on the specific scope and context of the deployment being proposed. What is considered unethical within one scope and context might be perfectly fine in another. I’ll illustrate with an example and then provide steps you can take to make sure your AI deployments stay ethical.

There are limited tests of fully autonomous, driverless cars happening around the world today. However, the cars are largely restricted to low-speed city streets where they can stop quickly if something unusual occurs. Of course, even these low-speed cars aren’t without issues. For example, there are reports of autonomous cars being confused and stopping when they don’t need to and then causing a traffic jam because they won’t start moving again.

We don’t yet see cars running in full autonomous mode on higher speed roads and in complex traffic, however. This is in large part because so many more things can go wrong when a car is moving fast and isn’t on a well-defined grid of streets. If an autonomous car encounters something it doesn’t know how to handle going 15 miles per hour, it can safely slam on the brakes. If in heavy traffic traveling at 65 miles per hour, however, slamming on the breaks can cause a massive accident. Thus, until we are confident that autonomous cars will handle virtually every scenario safely, including novel ones, it just won’t be ethical to unleash them at scale on the roadways.

Some Massive Vehicles Are Already Fully Autonomous – And Ethical!

If cars can’t ethically be fully autonomous today, certainly huge farm equipment with spinning blades and massive size can’t, right? Wrong! Manufacturers such as John Deere have fully autonomous farm equipment working in fields today. You can see one example in the picture below. This massive machine rolls through fields on its own and yet it is ethical. Why is that?

In this case, while the equipment is massive and dangerous, it is in a field all by itself and moving at a relatively low speed. There are no other vehicles to avoid and few obstacles. If the tractor sees something it isn’t sure how to handle, it simply stops and alerts the farmer who owns it via an app. The farmer looks at the image and makes a decision -- if what is in the picture is just a puddle reflecting clouds in an odd way, the equipment can be told to proceed. If the picture shows an injured cow, the equipment can be told to stop until the cow is attended to.

This autonomous vehicle is ethical to deploy since the equipment is in a contained environment, can safely stop quickly when confused, and has a human partner as backup to help handle unusual situations. The scope and context of the autonomous farm equipment is different enough from regular cars that the ethics calculations lead to a different conclusion.

Putting The Scope And Context Concept Into Practice

There are a few key points to take away from this example. First, you can’t simply label a specific type of AI algorithm or application as “ethical” or “unethical”. You also must also consider the specific scope and context of each deployment proposed and make a fresh assessment for every individual case.

Second, it is necessary to revisit past decisions regularly. As autonomous vehicle technology advances, for example, more types of autonomous vehicle deployments will move into the ethical zone. Similarly, in a corporate environment, it could be that updated governance and legal constraints move something from being unethical to ethical - or the other way around. A decision based on ethics is accurate for a point in time, not for all time.

Finally, it is necessary to research and consider all the risks and mitigations at play because a situation might not be what a first glance would suggest. For example, most people would assume autonomous heavy machinery to be a big risk if they haven’t thought through the detailed realities as outlined in the prior example.

All of this goes to reinforce that ensuring ethical deployments of AI and other analytical processes is a continuous and ongoing endeavor. You must consider each proposed deployment, at a moment in time, while accounting for all identifiable risks and benefits. This means that, as I’ve written before , you must be intentional and diligent about considering ethics every step of the way as you plan, build, and deploy any AI process.

Related Posts

A Paranoid Future Scenario For AI

LLMs Are More Than Generative AI

Is Our Education System Optimizing For The Wrong Outcomes?

AI & Data Science Predictions For 2024

Contact Info

Kennesaw Campus 1000 Chastain Road Kennesaw, GA 30144

Marietta Campus 1100 South Marietta Pkwy Marietta, GA 30060

Campus Maps

Phone 470-KSU-INFO (470-578-4636)

kennesaw.edu/info

Media Resources

Resources For

Related Links

• Financial Aid
• Degrees, Majors & Programs
• Job Opportunities
• Campus Security
• Global Education
• Sustainability
• Accessibility

470-KSU-INFO (470-578-4636)

© 2024 Kennesaw State University. All Rights Reserved.

• Privacy Statement
• Accreditation
• Emergency Information
• Report a Concern
• Open Records
• Human Trafficking Notice

MS in Data Science Graduate Wins Best Poster at NJBDA

The poster was on Jyoti Yadav’s thesis work titled Ai in TB Detection on Medical Big Data with Health and Educational Impacts

Posted in: Awards and Recognition , Data Science , Students

Jyoti worked on this project with her thesis committee, Dr. Varde , Dr. Liu and Dr. Antoniou . External contributors also included Dr. Lei Xie from CUNY Hunter NY (and Weill Cornell Medical College NY) as well. This research focused on analysis of the CODA TB Challenge Data Set.

Jyoti will be joining NYU, New York University Tandon School of Engineering , for a PhD program in Biomedical Engineering.