From Precision Agriculture to AI Safety: GVSU Computing Researchers Tackle Real-World Challenges
Published March 2, 2026 by Esther Djan
Recent research from Grand Valley State University’s College of Computing highlights how data science and trustworthy AI are being applied to solve complex, real-world challenges, from advancing precision agriculture in the United States to improving cybersecurity and AI safety. This work includes one journal publication released in January 2026 and two conference papers accepted for presentation in March, reflecting strong research momentum and continued collaboration between College of Computing faculty and graduate students.
A major highlight is the January 2026 journal publication by Kamrul Hasan and Dancun Juma, titled “Beyond Conventional Surveys: A Machine Learning Approach to Understanding Precision Agriculture Adoption in the U.S.” published in the International Journal of Agricultural and Bisystem Engineering (WASET). The study explores how machine learning can complement traditional survey methods by using historical USDA agricultural data to analyze technology adoption across U.S. states. Using normalization, trend analysis, correlation testing, and k-means clustering, the research identifies patterns linking agricultural sales growth with precision agriculture adoption. While not proving causation, results show strong alignment between data-driven indicators and survey-based adoption measures, demonstrating how scalable analytics can support policymaking, infrastructure planning, and workforce training.
Publication: https://publications.waset.org/10014362.pdf?utm_source=chatgpt.com
The research group also has two papers accepted at the 2nd IEEE Conference on Secure and Trustworthy CyberInfrastructure for loT and Microelectronics (SaTC 2026), held March 24-26, 2026 in Houston, Texas. The conference focuses on security and trust across loT, edge computing, and modern cyberinfrastructure.
The first paper, by College of Computing graduate student, Joyce Malicha and Kamrul Hasan, “Benchmarking the Effectiveness of AI-Driven Red Teaming Across Safety-Aligned Language Models,” evaluates how effective AI-assisted red teaming is at identifying risks in modern language models.
The second paper, by College of Computing graduate student, Hilda Ogamba, Kamrul Hasan, and Samah Mansour, “Zero-Shot Network Intrusion Detection Using Sematic Learning and CVAE Augmentation,” explores how semantic learning and generative augmentation can improve detection of new and previously unseen cyber threats.
Together, these publications demonstrate how modern machine learning can support real-world decision making from improving modernization strategies to strengthening digital security systems. They also highlight the strong role of graduate student collaboration in producing impactful, publishable research.