Benjamin Mensah Dadzie
PhD Researcher · Computer Science · IB/AP Mathematics Tutor · Sports Analyst
Benjamin
Mensah Dadzie
Concept Drift · Data Streams · Machine Learning · Statistical Analysis
PhD candidate at the University of Silesia in Katowice, researching novel methods for detecting and characterizing drift in non-stationary data streams. I bridge rigorous statistical theory with practical machine learning applications and I bring the same analytical lens to sports.
Research Focus
Drift Detection in Data Streams
Novel feature-rank based methods for identifying when the underlying distribution of a data stream has shifted — critical for adaptive ML systems.
Multi-Classifier Drift Analysis
Leveraging ensembles of classifiers to provide robust, feature-based drift detection in non-stationary environments.
Statistical Tests for Drift
Application of the Kolmogorov–Smirnov test and related non-parametric statistics to two-dimensional streaming data.
Drift Scoring & Visualization
Methods for scoring and visually communicating concept drift dynamics to support interpretable ML systems.
Recent Publications
Porwik, Orczyk, Wrobel, Dadzie · "A novel method for drift detection in streaming data based on measurement of changes in feature ranks"
Journal of AI and Soft Computing Research →Dadzie, Porwik · "Feature-Based Drift Detection in Non-Stationary Data Streams Using Multiple Classifiers"
Springer ACIIDS 2025 →