John Adejoh, AUST Faculty Publishes A Breakthrough Study on Credit Card Fraud Detection

The African University of Science and Technology (AUST) has recorded another major research milestone with the publication of a groundbreaking paper by Mr. John Adejoh, a faculty member and Ph.D hopeful in the  Computer Science department, which was published in Big Data and Cognitive Computing, a **Q1 Scopus-indexed journal (SCITE score: 9.8)*.

His research titled “An Adaptive Unsupervised Learning Approach for Credit Card Fraud Detection,” originated from a presentation in Professor Peter Onwualu’s Research Methodology class at AUST.

Mr. Adejoh’s model combines advanced Artificial Intelligence techniques—including Autoencoders, Self-Organizing Maps, and Restricted Boltzmann Machines—with a dynamic thresholding mechanism.

The result is a system that achieves 98% accuracy, a 97% F1-score, faster training times, and fewer false alarms compared to traditional methods.

“This research reflects AUST’s commitment to fostering innovation that solves real-world challenges,” Adejoh said.

With global credit card fraud losses projected to exceed $38.5 billion by 2027, AUST’s research output continues to position the university as a leader in cutting-edge solutions across Artificial Intelligence, Data Science, and Financial Technology.

Read the full paper here: MDPI Journal Link.

‌‌

‌‌