Assistant Professor · Applied AI & Machine Learning · Cybersecurity · Data Governance · Computing Education
University of Cincinnati · ORCID · Google Scholar · DBLP · LinkedIn
I am Isaac Kofi Nti, an Assistant Professor in the School of Information Technology at the University of Cincinnati. I design and evaluate responsible, explainable, secure, and resource-efficient artificial intelligence systems for consequential real-world settings.
My research connects applied machine learning with cybersecurity, health informatics, educational technology, finance, agriculture, energy systems, and data governance. Across more than 16 years of teaching and research, I have authored more than 60 peer-reviewed articles and developed open datasets, reproducible experiments, security laboratories, and adaptable computing curricula.
My goal is practical: build intelligent systems that people can understand, evaluate, secure, and use responsibly.
- Responsible and explainable AI: interpretable machine learning, SHAP, LIME, human-in-the-loop decision support, trustworthy AI, model transparency, and accountable deployment.
- AI for cybersecurity: ransomware and malware detection, memory forensics, intrusion detection, adversarial robustness, privacy-aware analytics, and secure data systems.
- Health and educational AI: clinical decision support, health informatics, predictive learning analytics, student-retention modeling, and ethical interventions.
- Efficient machine learning: compact neural architectures, resource-aware inference, synthetic data, structured feature selection, and evaluation under distribution shift.
- Data governance and reproducibility: provenance, privacy, research data management, reproducible workflows, systematic reviews, and bibliometric analysis.
- Computing education: cybersecurity laboratories, cloud learning environments, open educational resources, and responsible generative-AI integration.
| Year | Publication | Research contribution |
|---|---|---|
| 2026 | SHAP-guided feature refinement for efficient, robust, and interpretable memory-forensic malware detection | A memory-forensics pipeline combining SHAP-guided feature refinement with efficiency analysis, adversarial robustness testing, and family-disjoint validation. |
| 2026 | Hierarchical Sparse Neural Networks for Structure-Aware Ransomware Detection Under Distribution Shift | A taxonomy-aligned sparse neural architecture evaluated across temporal, family-disjoint, and open-set ransomware protocols. |
| 2026 | Synergistic Phishing Intrusion Detection: Integrating Behavioral and Structural Indicators with Hybrid Ensembles and XAI Validation | An explainable hybrid framework showing how behavioral, structural, and domain indicators jointly strengthen phishing detection. |
| 2026 | Explainable Machine Learning for Student Dropout Prediction and Tailored Interventions in Online Personalized Education | An explainable learning-analytics framework connecting student-risk predictions with human-reviewed intervention strategies. |
| 2026 | Interpretable Hybrid Three-Tier LSTM Model for Accurate and Transparent Breast Tumor Classification in Clinical Decision Support | A compact hybrid LSTM pipeline combining tumor classification with SHAP- and LIME-based explanations. |
For the complete and current publication record, use Google Scholar, ORCID, or DBLP.
| Repository | Type | Purpose |
|---|---|---|
| SynthDataHub | Research data | Citation-ready synthetic datasets for teaching, research, and responsible machine-learning experimentation across multiple domains. |
| 3-Tier-LSTM | Research software | Code and reproducible workflow supporting an interpretable hybrid three-tier LSTM for breast-tumor classification. |
| Information Security & Assurance Labs | Security education | Applied laboratories covering OpenSSL, TLS, SSH, VPNs, cryptography, and information-assurance workflows. |
| Introduction to Python for IT | Open curriculum | A browser-accessible Python foundation for learners from computing and non-computing disciplines. |
| Systematic & Bibliometric Literature Review | Research methods | Workshop materials, templates, scripts, and guidance for transparent evidence synthesis. |
| Data Technologies Administration Labs | Open curriculum | An enterprise data-architecture simulation covering design, governance, protection, validation, and defense. |
I welcome research and applied collaboration in:
- explainable and trustworthy AI for high-stakes decisions;
- secure and privacy-aware machine learning;
- ransomware, malware, and memory-forensics analytics;
- health informatics and responsible clinical decision support;
- learning analytics, student success, and AI-supported education;
- synthetic data, data governance, and reproducible computational research; and
- open cybersecurity and computing education.
Prospective collaborators, students, and research partners can begin with my UC research profile, review the artifacts above, or connect through LinkedIn.
My teaching and mentorship emphasize project-based learning, research integrity, ethical technology practice, and deployable technical skills. Areas include:
- Machine Learning and Data Mining
- Information Security and Assurance
- Principles of Cybersecurity
- Data Technologies Administration
- Fundamentals of Information Technology
- Graduate IT projects and applied research mentorship
Research methods: experimental design · predictive modeling · explainable AI · adversarial evaluation · systematic reviews · bibliometric analysis · reproducible research
Technical practice: Python · R · SQL · Jupyter · machine learning · deep learning · data visualization · Git · GitHub · cloud-based laboratories
Responsible deployment: data governance · privacy-aware analytics · model documentation · secure workflows · human oversight · open educational resources
- Full name: Isaac Kofi Nti
- Preferred scholarly name: Isaac Kofi Nti
- Role: Assistant Professor, School of Information Technology, University of Cincinnati
- ORCID: 0000-0001-9257-4295
- Scopus Author ID: 57210637914
- Web of Science ResearcherID: E-2004-2017
- DBLP: 260/5204
- Research profile: University of Cincinnati
- Machine-readable profile: research-profile.json
- Agent-readable summary: llms.txt
Responsible AI · Secure analytics · Reproducible research · Open computing education