College of Computing and Software Engineering

First-year scholars are tackling diverse challenges such as leveraging generative models to protect machine learning frameworks, enhancing security in AI-based healthcare systems, and developing human-AI teaming solutions for behavioral health. Guided by our expert faculty, these research projects aim to push the boundaries of computer science, ensuring technological advancements are both innovative and secure.

Explore the future of computing through research at KSU and its profound impact on today's fields!

Return to the Main Project Listings Page Questions: Email Us

Computer Science (Chen Zhao)

Enhancing Alzheimer's Disease Staging Prediction Through Multi-Modal Data Integration 

  • Alzheimer's disease (AD) is the most prevalent neurodegenerative disorder and the 6th leading cause of death in the United States. This project aims to enhance AD staging prediction by integrating structural MRI, genomics, cerebrospinal fluid (CSF) biomarkers, and electronic health records (EHR) using advanced deep learning models. Structural MRI is widely used for capturing brain changes in AD diagnosis. Genomic studies have identified single nucleotide polymorphisms (SNPs) correlated with brain structural changes in AD patients. Abnormal A-beta and tau protein levels in CSF are also highly correlated with AD. EHR data provides a three-stage classification: cognitively normal (CN), mild cognitive impairment (MCI), and AD. However, SNP expression does not differentiate between MCI and AD, and imaging data shows structural similarities between early MCI and CN, and late MCI and AD. Thus, integrating these four modalities can significantly improve AD staging prediction accuracy.

    Multi-modal learning promises higher predictive performance compared to single-modal approaches. However, direct concatenation of features from different modalities poses significant challenges in aligning multiple feature spaces during the fusion stage due to feature heterogeneity. This misalignment often results in poor model fusion performance. In this project, we will conduct comprehensive data processing for imaging, genetics, and EHR data using the Alzheimer's Disease Neuroimaging Initiative (ADNI) datasets. We will develop novel machine learning and deep learning models capable of integrating multi-modal data, focusing on aligning feature spaces and optimizing the fusion stage to enhance predictive accuracy.

    By integrating multi-modal data, we anticipate a significant improvement in the accuracy of AD staging predictions. This project will contribute to a deeper understanding of AD progression and support the development of more precise diagnostic and prognostic tools. Through innovative deep learning techniques, we aim to enhance the accuracy and reliability of AD staging, ultimately contributing to better patient outcomes and advancing the field of neurodegenerative disease research.

    1. Learn image processing algorithms, genetic data processing tools, and tabular data processing techniques.
    2. Apply fundamental machine learning and deep learning techniques for multi-view information fusion.
    3. Acquire skills in scientific paper writing and conference presentations.
    4. Gain interdisciplinary knowledge in both machine learning and bioinformatics.
    1. Conduct literature review and background study on AD staging prediction and multi-view machine learning.
    2. Process MRI/PET images, whole genome sequencing data, protein data, and EHR records.
    3. Develop novel multi-modality feature integration algorithms for AD staging classification.
    4. Prepare project reports and poster presentations for the annual symposium.
  • Hybrid
  • Dr. Chen Zhao, czhao4@kennesaw.edu

Computer Science (Arthur Choi)

Towards Bounding the Behavior of Neural Networks 

  • Recent and rapid advances in Artificial Intelligence (AI), particularly in the form of deep neural networks, has opened many new possibilities, but it has also brought with it many new challenges. In particular, it has become increasingly apparent that while deep neural networks are highly performant, they can also be opaque and brittle. We do not have enough understanding of why and when they work well, and why they may fail completely when faced with new situations not seen in the training data.

    Over the past few years, our research group has developed a symbolic approach to explaining and formally verifying the behavior of machine learning models.  More recently, we have developed an approach for bounding the behavior of individual neurons, based on a formal decomposition of a neuron into logical components. Subsequently, we can propagate these bounds across a network of neurons, in order to bound the behavior of a neural network. Our approach facilitates the analysis of a neural network, helping us to understand its behavior, and in turn, providing directions towards learning better models.

  • Real world experience working with (and developing) AI/ML models and tools, for preparation either for research in a graduate program (PhD), or for preparation for research/practice in AI/ML fields in industry.
    • AI/ML coding
    • AI/ML modeling and development
    • Reading papers
  • Face-to-Face
  • Arthur Choi, achoi13@kennesaw.edu

Computer Science (Kun Suo and Bobin Deng)

Building Highly-efficient and Low-power Artificial Intelligence Workloads on the Edges 

  • As edge and Internet of Things (IoT) technologies become increasingly popular, AI applications are more frequently being deployed on edge devices. However, these AI applications are often computationally intensive and lead to significant power consumption issues, which are at odds with the resource-constrained environment of edge devices. This project aims to address the performance and energy challenges of edge platforms through a new generation of power measurement instruments, along with learning, monitoring, and optimization techniques. Our focus is on studying resource management and energy consumption across various edge devices, with future plans to explore opportunities by creating data-driven runtimes and custom edge frameworks and data planes. The ultimate goal is to develop solutions for efficient and low-power edge computing and infrastructure.

    Specifically, this project will:

    1. Understand the technology and state-of-the-art edge computing platforms.
    2. Analyze the semantic gap between cloud and edge computing and design solutions to bridge this gap.
    3. Utilize advanced instruments to measure and understand the resource and energy consumption impacts of AI.
    4. Enhance the fundamental understanding of cloud-edge systems in terms of resource management and energy control.
    1. Attain an ability to design, implement, and evaluate edge computing system, process, component, or program to meet desired needs
    2. Gain system research skills through weekly meeting
    3. Gain knowledge in using IoT devices to run different applications with much efficiency and better performance
    4. Work collaboratively with other undergraduate and graduate students
    5. Improve the ability to solving real-world problem by self

  • Weekly group meeting and report the progress:

    1. Share the results on Microsoft teams with advisor in discussion
    2. Design and implement a computer-based system
    3. Evaluate the system performance using microbenchmarks or applications
    4. Present work on the C-day of College of Computing and Software Engineering
  • Hybrid

  • Dr. Kun Suo, ksuo@kennesaw.edu

    Dr. Bobin Deng, bdeng2@kennesaw.edu

Computer Science (Yong Shi)

Quantum Machine Learning for Cybersecurity and Science & Engineering Data

  • Machine Learning is known to provide solutions for data analysis and interpretation, and it is used in various fields such as computer vision, malware detection, and drug discovery. However, traditional machine learning approaches are incapable of successfully extracting useful information from large data sets, as they require tremendous time and resources while being performed on traditional computers. Quantum computing is a new type of qubit-enabled computing paradigm based on quantum properties such as superposition, interface, and entanglement for data processing and other tasks. It can be used to work on problems traditional supercomputers would not be able to handle efficiently. Quantum Computing can collaborate with Machine Learning for faster computation and more accurate data analysis, and Quantum Machine Learning (QML) has gained a lot of attention from both academia and industry recently.  

    Two of the most important applications of QML are cybersecurity and analyzing data from various science and engineering fields, such as biology and industrial engineering. This project will begin with learning Machine Learning and Quantum Computing, then followed by the development of a system in Python (developed in Google Colab) that uses the Quantum Tensorflow package and applies QML algorithms to process various (1) security and malicious data sets (2) science and engineering data sets and compares the performance with classical Machine Learning (CML) algorithms. In this project, the students will also conduct research on various quantum platforms such as Microsoft Quantum Development Kit, IBM Qiskit, Google AI Quantum Cirq, PennyLane, and more.

  • Major Work, Milestones and Expected Outcome:

    Stage 1. Basic understanding of Quantum Computing and Machine Learning 

    Stage 2. Source code and demo for a system that applies QML algorithms to process various (1) security and malicious data sets (2) science and engineering data sets and compares the performance with CML algorithms. And a Google site to host the hands-on learning modules in your application.

    Suggested modules (students can revise, after discussion with the professor) are:

    • Quantum Neural Network (QNN) / Deep learning Model
    • Quantum Random Forest Classifier
    • Quantum Support Vector Machine SVM 
    • Quantum Principal Component Analysis (QPCA) 
    • Quantum logistic regression 
    • Quantum Bayesian Network 
    • Quantum Decision Tree Classifier 
    • More

    The large security data sets used for those modules can be for:

    • DOS prevention 
    • Malware detection 
    • Finance Fraud detection 
    • Ransomware prevention and detection 
    • user behavior anomaly detection   
    • Spam email filtering 
    • Website Phishing prevention and detection
    • Predicting Product Backorder
    • Quality Prediction and Inspection
    • Patient Flow Analysis and Optimization Environments
    • Energy Consumption Estimation
    • Maintenance Action Recommendation
    • More

  • The students will perform:

    Task 1: Conduct research survey on various computing environments for our approach
    Task 2: Gather real world data sets to test our approach
    Task 3: Implement our algorithm and conduct experiments 
    Task 4: Collaborate with the advisor on a research paper
    Task 5: Help the advisor prepare for grant proposals for NSF programs

    The research work in this project will be submitted to several prestigious conferences such as IEEE INFOCOM: IEEE International Conference on Computer Communications, and the research work will be further extended for journals such as IEEE Transactions on Pattern Analysis and Machine Intelligence. The research work in this project will also be integrated to grant submission to NSF grant proposals for Research Experiences for Undergraduates (REU) (19-582), Improving Undergraduate STEM Education: Education and Human Resources (IUSE: EHR) (21-579), etc.

  • Hybrid
  • Dr. Yong Shi, yshi5@kennesaw.edu

Computer Science (Ahyoung Lee)

AI-Driven Water Quality Monitoring System Utilizing Experimental IoT Sensor Networks 

  • This project is focused on developing and implementing an AI-driven water quality monitoring system designed to predict waterborne bacteria levels in surface waters. The methodology includes the collection of water quality parameters—such as temperature, turbidity, pH, and other relevant factors—from surface waters using IoT sensor devices. These devices leverage a long-range, low-power wide-area network (LPWAN) protocol to ensure energy efficiency and reliable data transmission. The project also incorporates AI and machine learning algorithms to continuously monitor surface water quality and accurately predict bacteria levels.

    Interdisciplinary Research Collaboration:
    This is a multidisciplinary research initiative involving collaboration across various fields.

    (1) Computer Science Students: Focus on the development of AI/ML algorithms and mobile app design.
    (2) Engineering Students: Work on the hardware aspects, including microcontrollers and sensors.
    (3) Environmental Engineering Students: Contribute to the sustainability aspects of the project.
    (4) Biology Students: Handle bacterial contamination testing.


    Optional Research Activities:
    There is an opportunity to participate in hands-on water quality testing at Rottenwood Creek, the KSU Field Station, and possibly the Chattahoochee River. Participants may use waterproof rubber waders and boots to enter the water. Participation in these activities is optional.

  • At the end of the project, students should be able to:

    1. Conduct high-impact research to solve crucial computer network problems
    2. Analyze computer science, engineering, and data analytics approaches in the fields of sustainable and energy-efficient computing, be able to integrate software, hardware, communication systems, and network infrastructures
    3. Understand how to read academic papers in Computer Science
    4. Prepare and deliver a presentation demonstrating understanding of a paper or new tool or piece of technology
    5. Analyze results and shortcomings of published work
    6. Develop a research project and paper
  • Students will do research related to a research topic of computer networks, especially in the fields of sustainable and energy-efficient computing and computing for a more sustainable planet. Thus, I advise students to produce at least one manuscript submission targeting IEEE conferences or Journals per semester. 
  • Hybrid
  • Dr. Ahyoung Lee, alee146@kennesaw.edu

Computer Science (Kazi Islam and Md Abdullah Al Hafiz Khan)

Development of Trustable Machine Learning Models 

  • With the advancement of Artificial Intelligence (AI) technology, it is used in every part of our lives, e.g., remote sensing, healthcare, finance, security systems, autonomous vehicles, cybersecurity, transportation system, text generation, and human-AI Teaming. While AI has achieved state-of-the-art performance in these domains,  its black-box nature remains a critical challenge. Malicious actors can utilize adversarial attacks to access and manipulate the AI-based system. The attackers can use adversarial example attacks to craft a poison sample in the inference phase to predict incorrect information or perform a model poisoning attack to steal confidential data. To address these limitations, we must enhance trust in machine learning models.

    In this project, students will create trustable machine learning frameworks to make the model secure, explainable, confidential, and private. Students will explore the potential security vulnerability primarily in AI models, but this can be extended according to research findings. Students must read research papers, government websites, newspapers, and white papers to prove their hypothesis. Students will learn and apply machine learning algorithms and adversarial attacks to demonstrate their hypotheses empirically, starting from a foundational level. 

    • Investigate cutting-edge AI Technologies.
    • Learn machine learning algorithms and machine learning library, e.g., PyTorch and TensorFlow
    • Develop scientific writing skills by reading, critically analyzing, and discussing key scientific conference papers with mentors.
    • Improve presentation skills by participating in conference/workshop venue based on the research outcome.
    • Advance communication skills based on regular meeting with mentors.
    • Collaborate with other graduate/undergraduate students.
    • Attend weekly scheduled meeting in-person/online, present weekly outcome and receive mentor’s feedback.
    • Read research papers and summarize the finding in a scientific paper format.
    • Learn Machine learning library, e.g., PyTorch and TensorFlow
    • Execute and modify (if needed) a sample ML/AI source code and generate and analyze results.
    • Implement algorithm and execute experiments to get hands-on experience in cutting edge research.
    • Assist mentor writing research papers or submitting grants.
  • Hybrid

  • Dr. Kazi Aminul Islam, kislam4@kennesaw.edu

    Dr. Md Abdullah Al Hafiz Khan, mkhan74@kennesaw.edu

Computer Science (Md Abdullah Al Hafiz Khan, Abm Adnan Azmee, Francis Nweke, and Kazi Islam)

Human AI Teaming Utilizing Natural Language Processing

  • Artificial intelligence plays an important role in improving decision-making processes in real-world scenarios. However, in Human AI teaming, when people collaborate with AI, the partnership results in improved decision-making outcomes, particularly when dealing with complex and critical scenarios. On the other hand, AI also supports humans by offering data-driven insights, streamlining operations, and enhancing decision-making capabilities. This mutual relationship fosters understanding, effective communication, and more productive teamwork. For instance, AI-supported education involves both teachers and AI learning from each other to enhance student performance. Our study delves into how humans can improve AI systems and how AI can boost potential in return. By focusing on text-based interactions, we aim to create machine learning algorithms that enhance communication synchronization for reliable and sustainable decision-making processes. 

    In this research project, we aim to explore efficient human-AI teaming technologies and design prototype systems for them. Our research team has an extensive background in developing machine learning, deep learning, and large language models (LLMs). Given our current skills and expertise, we plan to explore more with this project.

    •  Develop scientific literature review skills by reading and analyzing technical articles and blogs.
    • Real-world experience developing software applications, AI/ML/LLM models, and tools for preparation for research in AI/ML in the industry.
    • Improve communication and scientific writing skills.
    • Present research results.
    • Reading papers and technical articles
    • Develop software applications and annotate data.
    • Execute and modify (if needed) a sample ML/AI/LLM source code and generate and analyze results.
    • Meet weekly (in-person or virtually) and report progress.
  • Hybrid

  • Dr. Md Abdullah Al Hafiz Khan,  mkhan74@kennesaw.edu

    Abm Adnan Azme, aazmee@students.kennesaw.edu 

    Francis Nweke, fnweke@students.kennesaw.edu

    Dr. Kazi Aminul Islam, kislam4@kennesaw.edu

Computer Science (Xinyue Zhang)

Utilizing AI to Understand the Impact of Social Determinants of Health on Alzheimer's Care Equity 

  • Alzheimer's disease, the most common form of dementia, will rise from 6.5 million to approximately 13.8 million by 2060, surging by 4.7% in the U.S. Eighty-three percent of the Alzheimer patients are cared for at home by informal caregivers (family and friends), totaling over 11 million Americans who provided an estimated 18.4 billion hours of unpaid care in 2023 which is valued at 346.6 billion. With 83% of people with Alzheimer’s disease and related dementias (Pw-ADRD) living in their own homes, informal caregivers serve as mediators for symptoms, sources of information, and emotional support. However, many informal dementia caregivers continue to lack the knowledge and skills to manage behavioral symptoms effectively. Therefore, significant disparities persist in the quality and accessibility of Alzheimer's care, particularly among underserved populations.

    These disparities are mostly because of the social determinants of health (SDOH) such as income, education, and healthcare access, which has a significant impact on the care patients receive and their health outcomes. In this project, we aim to leverage AI models to analyze the impact of SDOH on Alzheimer's care equity. We propose to apply both supervised and unsupervised learning models on Alzheimer's Disease and Healthy Aging Data published by CDC to identify the correlations between the SDOH and care equity. The findings from this project can provide invaluable insights into the care disparities problem in Alzheimer’s disease. 

    1. Apply fundamental and disciplinary concepts and methods that support the research project.
    2. Attain the ability to identify, analyze, and solve problems creatively.
    3. Investigate the cutting-edge AI models.
    4. Propose solutions to deal with real-world datasets.
    5. Learn the principles of academic writing and research presentation skills.
    6. Collaborate with other graduate and undergraduate students with effective oral and written communication.
    1. Weekly meetings and updates.
    2. Implement the state-of-the-art AI models.
    3. Explore the literature to understand the health equity problem in Alzheimer's disease.
    4. Prepare presentations for key findings in the project.
    5. Final research project reports and poster presentation in the annual symposium
  • Hybrid
  • Dr. Xinyue Zhang, xzhang48@kennesaw.edu

Compuer Science (Zongxing Xie)

A Study to Explore COTS Wireless Sensor for Smart Health Applications 

  • In this project, we will use commercial-off-the-shelf wireless sensors to collect data about the physical environment, where human subjects perform daily activities. The idea is when wireless signals will interact with and bound off the physical environment including human subjects. Different movements will result in different signal patterns. We will analyze the collected data to discover data patterns and understand their correlations with human activities using AI tools, such as machine learning, deep learning, and data visualization. We will investigate the opportunities and challenges in leveraging the discovered data patterns and correlations for healthcare applications, such as sleep quality assessment, gait pattern analysis, fall detection and fall risk prediction.
    • Students will develop research skills, such as surveying literature, identifying problems, designing experiments, analyzing results, presenting findings and insights.
    • Students will gain hands-on experience with sensor systems.
    • Students will gain hands-on experience with cutting-edge AI technologies.
    • Students will develop a good understanding of methodologies for conducting scientific research.
    • Students will develop teamwork and leadership skills from this project.
    • Students should meet with the advisor weekly, during which students should communicate their progress with the advisor.
    • Students are expected to summarize their results and the status of the planned action items. 
    • Students are expected to discuss and identify the planned action items (e.g., literature review, ideation, experiment design, algorithm implementation, evaluation and analysis) for the next week.
    • Students are expected to reach milestones in a timely manner.
  • Hybrid
  • Dr. Zongxing Xie, zxie1@kennesaw.edu

Data Science and Analytics (Kevin Gittner and Lauren Matheny)

Fraudulent Large Language Model (LLM) Response Detection in Incentivized Online Survey Data Collection 

  • In the world of research, online surveys have become a vital tool for gathering information. However, these surveys are increasingly targeted by both human fraud actors and automated bots that generate fake responses, compromising the quality and reliability of the data collected. This project aims to tackle this growing problem by developing advanced methods to detect and prevent fraudulent responses in online surveys.

    Our Data Quality Survey Methods Lab is at the forefront of this effort. We specialize in identifying and addressing issues related to the quality of online survey data. Our research focuses on public health and policy, where accurate data is crucial for informed decision-making. We have been actively developing innovative techniques to enhance data quality and ensure that survey results are reliable and trustworthy.

    One of our key projects involves creating an automation script that mimics human responses to surveys using Large Language Models (LLMs) like ChatGPT, Gemini, LLaMa2, etc. By generating large datasets of known LLM responses, we aim to provide valuable tools for researchers to test and refine methods for detecting fraudulent survey responses. We are developing an automation script that can handle various types of survey questions. This script will then be extended to automate multi-page surveys, ensuring a smooth and efficient process. By integrating localized LLMs, we are refining techniques to match LLM-generated content with survey response options while also developing and testing various methods to detect poor quality and fraudulent data, which will be extended to API based LLMs.

    Our projects emphasize collaboration and mentorship. We work closely with students from undergraduate to PhD. levels, providing them with hands-on experience in cutting-edge survey methodology, data quality control, and LLM and AI automation scripts. This collaborative environment fosters innovation and prepares students for future careers in data science and analytics.

    By combining our expertise in online survey data quality with advanced automation techniques, we aim to create robust tools that can detect and prevent fraudulent survey responses. This project not only addresses a critical issue in research but also contributes to the development of practical solutions that can be applied across various fields, ensuring that the data used to inform decisions is accurate and reliable. Join us in this exciting journey to enhance the integrity of online survey data and make a meaningful impact in the world of research.

  • Technical Skills
    Coding Proficiency: Learn and enhance programming skills, particularly in Python, for developing and debugging automation scripts.
    Data Analysis: Gain experience in coding (in R and SAS) and analyzing large datasets to identify patterns, ensure accuracy, and evaluate survey data quality.

    Research Skills
    Literature Review: Develop the ability to conduct comprehensive literature reviews on survey methodologies, LLM applications, and fraud detection.
    Experimentation: Design and conduct experiments to test and refine the automation script and LLM integration.

    Analytical Skills
    Problem-Solving: Strengthen problem-solving abilities by identifying and troubleshooting errors in coding.
    Statistical Analysis: Learn to apply statistical methods to analyze real world data for various purposes.

    Collaboration and Communication
    Team Collaboration: Improve teamwork skills by working closely with peers and mentors on various research tasks.
    Presenting Research: Gain experience in preparing and delivering presentations and writing research papers for academic dissemination.

    Project Management
    Documentation: Develop skills in maintaining detailed records of research progress, including code changes and data analyses.
    Time Management: Learn to manage time effectively by balancing multiple tasks and meeting project deadlines through a hybrid work environment.

    Professional Development
    Mentorship: Benefit from regular guidance and feedback from experienced researchers.

    Innovative Thinking
    Creative Inquiry: Engage in creative problem-solving to develop and refine techniques for detecting poor quality and fraudulent data in surveys.
    Adaptability: Learn to adapt to new challenges and technologies in the rapidly evolving field of data science and survey research.

  • Role adaptability is necessary as weekly duties will frequently transition. Some various duties might include:

    Literature Review: Stay updated on the latest research in survey methodologies, LLM applications, and fraud detection techniques. Will require extensive reading and Q&A to gain an understanding of the "why" of our research.

    Data Collection and Analysis: Analyze data from previously collected rounds of data. Design new surveys for data collection.

    Coding and Development: Review, write, and debug code for the automation script and statistical analyses. Learn about integrating various Local Language Models (LLMs). 

    Team Meetings: Participate in weekly meetings to discuss progress and receive feedback.

    Testing and Quality Assurance: Test the proof of concept automation script. Test novel methods for fraud and LLM detection. Identify and troubleshoot errors.

    Documentation: Maintain detailed records of code changes, test results, and data analyses.

    Mentorship and Collaboration: Regular check-ins with mentors for guidance. Collaborate with peers on research tasks.

    Research Dissemination: Prepare findings for presentations and publications.

  • Hybrid

  • Dr. Kevin Gittner, kgittner@kennesaw.edu

    Dr. Lauren Matheny, lmathen1@kennesaw.edu

Information Technology (Honghui Xu)

Privacy-Preserving Multimodal Sentiment Analysis 

  • Our project focuses on multimodal sentiment analysis, which combines data from various sources—like text, images, and audio—to understand public sentiment. This cutting-edge approach can predict outcomes in the stock market and elections, assess customer satisfaction, and enhance interactions between humans and computers.

    Currently, as deep learning technology advances rapidly, using these sophisticated methods for sentiment analysis has become increasingly effective. However, this poses privacy risks. For example, the detailed data we gather could unintentionally reveal personal information such as someone’s identity or location, leading to privacy breaches.

    To address these concerns, our project develops new ways to protect privacy while still making the most of multimodal data for sentiment analysis. We’re tackling challenges like high data correlation, which can weaken privacy protection methods like differential privacy. Differential privacy usually works by adding noise to the data, but this can become complicated when different types of data are interconnected.

    Our goal is to create a system that not only protects privacy but also maintains the utility and accuracy of the data analysis—ensuring that as we enhance technological capabilities, we also safeguard individual privacy.

    This project is perfect for students interested in cutting-edge technology and real-world applications. They’ll gain experience in both developing new technologies and understanding their broader implications, including privacy issues. Join us to contribute to research that sits at the forefront of technology and society.

    1.  Multimodal Data Handling: Students will learn how to manage and analyze data that integrates multiple forms of media, such as text, images, and audio. This includes preprocessing, normalization, and the extraction of meaningful features from each modality, providing a foundation for handling complex data structures.
    2. Privacy-Preserving Techniques: The core of the project involves the application of differential privacy and adversarial training methods. Students will learn how to implement these techniques to protect user data from potential privacy breaches. This includes understanding how to add noise to data in a way that maintains its utility while safeguarding sensitive information.
    3. Machine Learning and Deep Learning: Participants will gain hands-on experience with machine learning algorithms, focusing on how they can be applied within multimodal AI systems. This includes training generative models and using neural networks to perform sentiment analysis, enhancing their understanding of AI’s potential and limitations.
    4. Critical Thinking and Problem Solving: By tackling the challenge of data correlation and its impact on privacy, students will enhance their ability to think critically about the interplay between different data types and the consequences of data processing decisions.
    5. Research Methodology: Students will engage in the entire research process, from hypothesis formation and literature review to experimentation and analysis. This comprehensive exposure will help them understand how to design studies, interpret results, and present findings clearly and effectively.

  • Fall 2024:

    1.  Weeks 1-4: Orientation and Training
       1) Introduction to project goals, relevance, and expected outcomes.
       2) Training sessions on data handling, privacy principles, and basic machine learning concepts.
       3) Familiarization with software tools and programming languages necessary for the project (e.g., Python, TensorFlow).
    2.  Weeks 5-8: Data Preprocessing and Analysis
       1) Hands-on activities involving the collection, cleaning, and preprocessing of multimodal data.
       2) Initial analysis of datasets to understand underlying patterns and correlations.
       3) Weekly meetings to discuss findings and troubleshoot issues.
    3.  Weeks 9-12: Introduction to Privacy-Preserving Techniques
       1) Training on differential privacy and adversarial methods.
       2) Implementation of basic privacy-preserving algorithms on small datasets.
       3) Group discussions with PI's graduate students on the implications of privacy techniques on data utility and integrity.

    Winter Break: Students are encouraged to review literature related to their project work, focusing on recent advances in privacy-preserving technologies and their applications in AI.

    Spring 2025:

    1.  Weeks 1-4: Advanced Implementation
       1) Application of advanced privacy-preserving techniques to larger, more complex multimodal datasets.
       2) Regular sessions with mentors to refine methods based on feedback.
    2. Weeks 5-8: Evaluation and Optimization
       1) Systematic evaluation of models for effectiveness and privacy protection.
       2) Optimization of models based on evaluation outcomes and mentor feedback.
    3. Weeks 9-12: Finalization and Presentation
       1) Preparation of final reports detailing methodologies, results, and learning experiences.
       2) Development of presentations to share findings with peers, faculty, and possibly at student research symposiums.
       3) Reflection on the project and discussion of potential future work or improvements.
  • Hybrid
  • Dr. Honghui Xu, hxu10@kennesaw.edu

Information Technology (Liang Zhao)

Website Fingerprinting: Attacks and Defenses

  • Website fingerprinting acts like a detective trying to guess what you've been up to online. Imagine someone looking at the digital 'footprints' left by your web browsing - the timing, direction, and size of the data you send and receive. Even if you use tools to keep your online activities private, like proxies, VPNs, or Tor, this detective could potentially piece together which websites you've visited. In this field, we use machine learning (think of it as a smart computer program that can learn from patterns) to better understand and protect online privacy. In the case of website fingerprinting, these smart programs can analyze your data footprints and guess the websites they originated from.

    In this project, we will study how to safeguard digital privacy or understand the tactics used to compromise it through an application of web security. 

    1. Improve research and technical skills.
    2. Develop literature reviews about machine learning for cybersecurity.
    3. Gain understanding and knowledge about computer networking.
    4. Implement web applications using real-world data.
    5. Improve communication skills and scientific writing skills.
    1. Read papers to gain knowledge about computer networking and cybersecurity.
    2.  Literature review and report of findings regarding machine learning for cybersecurity.
    3. Develop a prototype for testing the proposed attacks and defenses.
    4. Attend weekly meetings and report weekly updates.
  • Hybrid

  • Dr. Liang Zhao, lzhao10@kennesaw.edu

Information Technology (Chloe Xie)

AI-Driven Study of Tau Protein in Alzheimer's and Related Dementias (ADRD) 

  • This project focuses on utilizing cutting-edge computational methods, including data science and artificial intelligence, to study the tau protein's role in Alzheimer's Disease and Related Dementias (ADRD). The project aims to uncover new insights into the molecular mechanisms and potential therapeutic targets associated with tau pathology. By integrating large-scale data analysis and machine learning, the project seeks to advance our understanding of tau protein's involvement in neurodegenerative diseases, ultimately contributing to the development of novel treatment strategies.

  • Computational and Analytical Skills: Students will learn to use advanced computational tools and software for data analysis, including bioinformatics platforms, statistical software, and machine learning algorithms. They will gain proficiency in handling large datasets, performing data cleaning, visualization, and interpretation.

    AI and Machine Learning Techniques: The project will expose students to AI and machine learning methods, particularly in the context of biomedical data. They will learn to develop and apply predictive models, perform clustering, and use neural networks to identify patterns and insights related to tau protein and neurodegenerative diseases.

    Research Methodology: Students will be trained in rigorous scientific research methods, including hypothesis formulation, experimental design, and statistical analysis. They will learn how to critically evaluate scientific literature, identify research gaps, and develop research proposals.

    Bioinformatics and Molecular Biology: The project will provide students with a deeper understanding of bioinformatics, including sequence alignment, structural biology, and the analysis of protein-protein interactions. They will also gain insights into the biological mechanisms of tau protein and its implications in Alzheimer's Disease and Related Dementias (ADRD).

    Communication and Collaboration: Students will enhance their communication skills by presenting their findings through oral presentations, posters, and written reports. They will also learn to collaborate effectively within a multidisciplinary team, honing their ability to work with peers and mentors from diverse scientific backgrounds.

  • Data Collection and Cleaning: Collect and clean datasets related to tau protein, ensuring the data is ready for analysis.

    Computational Analysis: Use bioinformatics tools and programming languages like Python to perform key analyses, including machine learning modeling.

    Literature Review: Review and summarize relevant scientific literature to stay informed about recent advances and contextualize their research.

    Weekly Meetings: Participate in discussions with mentors and peers to review progress, share insights, and address challenges.

    Documentation: Prepare concise reports summarizing their findings and methodologies.

  • Hybrid
  • Dr. Chloe Yixin Xie, yxie11@kennesaw.edu

Information Technology (Taeyeong Choi and Sungchul Jung)

Feed the World! Developing VR Environments to Foster Future Agri-Food Experts 

  • With the global population projected to reach 9.7 billion by 2050, food production must double to meet future demand. The agrifood sector faces a critical labor shortage, worsened by declining interest from younger generations. In this project, student scholars aim to address this challenge by developing virtual reality (VR) environments that offer engaging, educational experiences in the food supply chain.

    To be specific, student scholars will design VR simulations covering all stages of food production—from farming, harvesting, and post-harvesting to cooking. In the VR environments, for example, gamers will experience tasks such as sowing seeds, watering plants, and identifying ripe crops, followed by post-harvest activities like sorting and packaging. In addition, they will engage in cooking simulations, following recipes like tomato pasta. By providing a fun and immersive learning platform, this project aims to spark interest in careers within the agrifood industry.

    Upon completion, promising extensions may also be considered to create a high impact on society. For instance, by collaborating with local high schools, the created VR content could be provided as learning materials for youths; VR games may be developed to host online competitions; and/or in-depth research could be conducted to assess the impact of the VR environments on shifting user interests and impressions in agrifood careers.

    Through this project, students will gain access to state-of-the-art VR technologies and acquire relevant technical abilities that will be transferable to their own environments. Moreover, the interdisciplinary nature of the proposed project will provide them an opportunity to learn specifically about real-world problems, potential technical solutions, and their impacts, and how they are interconnected.

  • Student scholars will eventually learn essential skills in computer graphics, game engines, and coding. In particular, Unity 3D will be a primary playground for the students to transform conceptual matters into more substantial, visual, and interactive objects. Moreover, as they utilize VR headsets as external tools, they will learn how to develop software to interface between a PC and the headsets for users' best experience. Moreover, C# will be used as the main programming language, with which the students will internalize various logics and notions that will be easily transferrable to other programming environments in a different language. 

    In addition, student scholars will be trained to hone communication skills. While presenting the weekly updates to the PIs, they will learn how to explain and visualize their work, ask questions, and respond to feedback. Students will also gain various skills for problem solving. PIs will help them decouple complex problems into sub-problems and identify actual issues. Through this process, they will be able to set up the plans for the next steps to make the progress. 

    Lastly, students will learn to connect computer science skills to social problems. They may not have thought of agriculture and food as the application domains of computer science before, but this project will provide them an opportunity to consider it and feel self-efficacy as a computer scientist who could contribute to the society. In other words, this project will broaden their view for the application of their engineering skills, and it will positively impact their academic trajectory in the following years.

  • They are encouraged to attend an in-person meeting, where any updates, challenges, questions, and needs are presented to the PIs. All the codes and visualizations are uploaded to an online repository so that all engaged researchers can access them. 

    Specifically, students will build up VR environments to reflect specific food production scenarios. They will configure the environments using 3D assets on the Unity 3D game engine, coding in C# for interaction features if needed. 

    In an initial phase, they will be advised to follow particular materials and interact with PI's graduate student research assistants to learn about the development environments. Afterward, they will use creativity to construct 3D VR environments for simulating food supply chains under PIs' supervision.

    While all these tasks can be performed remotely, if they desire, they can use the lab spaces and PCs there. For troubleshooting, they can freely contact PIs and the graduate students anytime. 
     

  • Hybrid

  • Dr. Taeyeong Choi, tchoi3@kennesaw.edu

    Dr. Sungchul Jung, sjung11@kennesaw.edu

Information Technology (Maria Valero de Clemente)

Creating Mobile App for Blood Donation 

  • The project aims to address the critical need for increased blood donation by leveraging mobile technology to streamline the donation process, improve donor engagement, and enhance overall accessibility. The mobile app will serve as a comprehensive platform that connects blood donors with appointments and reminders making the donation process more efficient and user-friendly.

    The app will have an intuitive design, making it easy for users to navigate and access all features with minimal effort. Donors can schedule appointments at their convenience, with real-time availability updates from nearby blood banks. Donors can track their donation history, view eligibility for their next donation, and receive notifications when they are eligible to donate again. The app will offer information on the importance of blood donation, eligibility criteria, and tips for first-time donors.

    This is a long-term project, the student will start with the design of the app and the mockup screens of how the app will look like. Eventually functionality will be integrated, but not all the functionalities will be required during the first year.

    1. Students will gain proficiency in programming languages commonly used in mobile app development.
    2. Students will learn how to design intuitive and user-friendly interfaces, focusing on layout, navigation, and user interaction principles.
    3. Students will develop skills in integrating various APIs
    4. Students will develop skills in writing and executing test cases to ensure that individual components and the entire app function as intended.
    5. Students will enhance their ability to identify challenges, think critically, and develop innovative solutions during the app development process.
    6. Students will improve their ability to communicate technical concepts to both technical and non-technical audiences, whether through presentations, reports, or discussions with stakeholders.
    1. Weekly meetings with faculty and graduate students
    2. Literature review
    3. Requirements gathering
    4. Familiarization with development environments
    5. Peer-review sessions
    6. Development of the app
    7. Documentation
  • Hybrid
  • Dr. Maria Valero de Clemente, mvalero2@kennesaw.edu

Information Technology (Nazmus Sakib and Syeda Salma)

Leveraging AI to Address Unregulated Screentime and Promote Healthy Development in Young Children

  • The pervasive presence of screens in today's world has led to heightened concerns about excessive and unregulated screen time, particularly in young children, potentially impacting their cognitive, emotional, and physical development. To tackle this issue, we propose leveraging AI to develop an evidence-based mHealth solution. We will create a smartphone-integrated system to monitor and analyze the effects of excessive screen time on young children. 

    To achieve this, we need to unpack data to identify parental challenges in managing screen time so that we can offer personalized recommendations to promote healthy digital habits. By leveraging AI and Machine Learning algorithms, we aim to provide tailored suggestions considering each parent-child dyad's digital lifestyle and socioeconomic factors. Our data-driven insights will empower parents to regulate their child's screen time effectively, promoting a balanced approach to digital engagement. Furthermore, we will conduct a thorough literature review to identify gaps in existing research and explore personal and socioeconomic factors influencing screen time, strengthening the foundation of our solution. Our goal is to provide a practical tool that supports healthy cognitive and emotional development in young children. 

    1. Enhance their scientific literature review abilities by engaging with and evaluating technical research journals and articles.
    2.  How to investigate, explore and address an issue with a proper writing skill
    3. Engage in implementing the algorithm and conducting experiments to gain practical experience in the latest research advancements.
    4. Hands on experience on Machine Learning implementation
    5. Gain the capacity to carry out a research-driven project while enhancing interpersonal skills like presentation abilities and exposure to a research setting. 
  • Students will engage in reviewing and evaluating scientific literature, developing research writing skills, implementing algorithms and conducting experiments, gaining hands-on machine learning experience, and presenting their research findings.
  • Hybrid

  • Dr. Nazmus Sakib, nsakib1@kennesaw.edu

    Syeda Umme Salma, ssalma@students.kennesaw.edu

Information Technology (Ying Chen)

AI-Empowered Intelligent Augmented and Virtual Reality Systems 

  • By offering immersive and interactive experiences, augmented reality (AR) and virtual reality (VR) are transforming the way we communicate, learn, play, and work and revolutionizing a wide range of industries, notably e-commerce, education, entertainment, manufacturing, and healthcare. In this project, we will enhance AR and VR systems with intelligent capabilities that improve system functionality and user experience. We plan to achieve this by integrating advanced artificial intelligence, specifically computer vision models and large language models (LLMs), into AR and VR systems. We will leverage these AI technologies to generate text descriptions that incorporate context information from both AR and VR users and their surrounding environments. The end goal of this project is for AR and VR systems to understand and respond to users' contexts and environmental conditions in real-time, creating a more immersive, interactive, and highly personalized experience tailored to user preferences and behavior patterns.

  • 1. The student will work directly with the PI, who will guide them in developing both a deep understanding of specific technical details, and a broader perspective on the associated areas.

    2. The student will collaborate and exchange ideas with graduate students in the lab.

    3. The student will hone their writing and presentation skills by presenting posers in the annual symposium and writing research reports. They will also have the opportunity to present their work at one of our lab meetings, allowing them to receive feedback from other lab members.

    4. The students will gain the following skills when conducting the research:

    1) Unity3D development
    2) virtual and augmen

  • Student weekly duties include:

    1) attending lab group meetings and/or one-on-one meeting with the PI

    2) reading existing papers in the field of AI-empowered AR and VR

    3) implementing prototypes using commercial AR and VR devices such as VIVE XR Elite, as well as AR and VR development platforms such as Unity

    4) writing reports that detail the experimental settings, summarize research findings, and provide insights from the results

  • Hybrid
  • Dr. Ying Chen, ychen62@kennesaw.edu

Software Engineering and Game Development (Sungchul Jung)

Immersive Learning and Training Using Extended Reality (XR) 

  • In this project, students will conduct research to understand the utilization of immersive technology in learning and training. They will employ virtual and mixed-reality headsets, delving into the realistic virtual avatar, spatial audio, and responsive visual interfaces within immersive environments compared to the non-immersive learning platform. Based on this exploration, the research will focus on applying cutting-edge technology to enhance human perception, cognition, attention, and empathic responses in learning and training. Participants will have the opportunity to contribute to the advancement of knowledge in the field of empathic immersive experiences. This will involve fostering innovative solutions for real-world challenges, all while developing a strong understanding of both theoretical investigations and practical implementations.
  • This project expects students to conduct high-impact research to address critical issues in cognition, perception, attention, and emotion in immersive experiences. Also, developing prototypes for research projects utilizing immersive technology tools, including Virtual/Mixed Reality devices within Unity or the Unreal game engine or similar, is critical. Student needs to prepare and deliver presentations to showcase their comprehension of academic papers, new development tools, sensors, or emerging technologies, including analyzing the outcomes and limitations of published research.
  • Weekly duties include attending lab meetings and individual meetings and submitting reports on scientific research work involving the implementation of prototypes using tools such as Unity or Unreal Engine.
  • Hybrid
  • Dr. Sungchul Jung, sjung11@kennesaw.edu

Software Engineering and Game Development (Lei Zhang and Chloe Xie)

Collaborative Learning of Complex Molecular Biology Concepts with a Multi-user Virtual Reality Storytelling Experience 

  • Complex molecular biology concepts such as DNA damage and repair mechanisms involve an intriguing process with different types of proteins and their interactions and present a learning challenge to students at all levels with traditional instructional approaches. Recent practices of combining immersive technologies such as virtual reality (VR) and digital storytelling to explain complex science concepts through an embodied and experiential learning experience provide promising new pedagogical strategies to supplement textbook instructions. Our previous research on a single user VR storytelling learning experience has shown increased engagement and motivation in learning target science concepts. We hypothesize that a multi-user setup will further increase learner engagement by allowing the students to take on different roles of story characters and having them explore and learn through shared narratives in virtual environments.

    This research project utilizes the latest VR technologies and devices to develop a multi-user immersive and interactive storytelling experience centering around the main DNA repair mechanisms: the roles of MRN complex and p53 protein molecules. The immersive story breaks down the complexity of the concepts with engaging sci-fi narratives and reinforces learning through gameplay interactions and role-playing. The overarching goal of the project is to explore how role-playing and scenario-based learning in the virtual environments can help with a learner’s comprehension of complex molecular biology concepts. 

    1. Research capability of doing literature review, user study data collection and analysis.
    2. Software development skills of creating immersive learning experiences in Unity for complex science learning
    3. Interactive design skills in UI/UX.
    4. 3D modeling skills to create digital assets for virtual environments.

  • Students will work on assigned project prototype development work and give weekly reports of the progress.

    They will also do a literature review on the latest research related to the project topic and prepare materials for presentation.

  • Hybrid

  • Dr. Lei Zhang, lzhang24@kennesaw.edu

    Dr. Chloe Yixin Xie, yxie11@kennesaw.edu