Muhammad Abusaqer

1) Security-Aware Language Intelligence

Incident-centric discourse triage and prioritization for cybersecurity events—especially healthcare cyber incidents—using transparent scoring and analysis workflows.

Example: CyberTweetGrader&Labeler (CTGL) for prioritizing cyber-incident tweets.

2) Comparative Benchmarking: ML ↔ Transformers ↔ LLMs

Controlled, reproducible comparisons that quantify not only predictive performance, but also reliability and deployment trade-offs (cost, latency, and failure modes).

Example: prompted LLM inference vs. fine-tuned encoders under constraints.

3) Green NLP & Efficient Inference

Energy-aware evaluation that supports sustainable model selection—e.g., TF-IDF baselines vs. fine-tuned transformers—without sacrificing responsible reporting.

Example: accuracy–latency–energy trade-offs for abuse detection.

Lab Director

Muhammad Abusaqer (Minot State University)

Security-aware NLP • trustworthy evaluation • deployment-realistic benchmarking

Students

The lab actively mentors undergraduate and M.S.-level research projects aligned with the thrust areas above.

(A current member list can be added/updated anytime.)

Former Undergraduate Researchers (Coauthors)

Grouped by year; initials are used to match publication records.

MICS 2025

• A. Pun — Predicting Student Academic Performance: Using Machine Learning and Clustering ( PDF) [Student success]
• B. Olson — Predicting Student Academic Performance: Using Machine Learning and Clustering ( PDF) [Student success]
• D. Degele — Analyzing Ransomware Incidents in Healthcare: Patterns and Risk Assessment ( PDF) [Healthcare ransomware]
• T. Khan — Evaluating Quick-Commerce Platforms: A Sentiment and Topic Modeling Analysis of User Reviews ( PDF) [Sentiment & topic modeling]

MICS 2024

• J. Jensen — Global Echoes of the FIFA World Cup 2022: Sentiment and Theme Analysis via Deep Learning and Machine Learning on Twitter [Event sentiment/theme]
• S. Khan — Text Detection between an AI-Written Passage vs. a Human-Written Passage [AI vs human text]
• K. Khan — Text Detection between an AI-Written Passage vs. a Human-Written Passage [AI vs human text]
• T. Smith — Predicting Campus Crime Based on State Firearm Policy [Public safety & policy]

MICS 2023

• C. Fofie — Cyberbullying Classification Using Three Deep Learning Models: GPT, BERT, and RoBERTa [Cyberbullying]
• Q. Sullivan — Darknet Traffic Classification Using Deep Learning [Network traffic]
• A. Scott — Automated Categorization of Cybersecurity News Articles through State-of-the-Art Text Transfer Deep Learning Models [Cybersecurity news]
• J. T. Snow — Automated Categorization of Cybersecurity News Articles through State-of-the-Art Text Transfer Deep Learning Models [Cybersecurity news]