Congratulations Otu Kumi Michael

Otu Kumi Michael is a graduating MPhil student from the Department of Computer Engineering at the Kwame Nkrumah University of Science and Technology (KNUST). His research interests include machine learning, artificial intelligence, cloud computing, predictive analytics, and explainable AI (XAI).

He developed a novel explainable machine learning framework for his MPhil thesis for enhancing task failure prediction in cloud computing environments using SHapley Additive exPlanations (SHAP). His research investigated how explainable learning techniques can improve transparency, interpretability, and accuracy in predicting task failures within large-scale cloud infrastructures. The proposed XGBoost-SHAP model demonstrated superior performance compared to conventional models such as Logistic Regression, Random Forest, K-Nearest Neighbors, and Decision Tree, setting a new benchmark for reliability and explainability in cloud-based task management.

His work highlights how integrating explainability into predictive models enables data-driven insight into the causes of task failures, helping cloud service providers optimise resource allocation and minimise operational costs. The research also emphasises feature contribution analysis using SHAP to identify the most influential parameters driving task failures, providing a practical framework for proactive cloud management.

The KNUST Engineering Education Project (KEEP) has been instrumental in his academic success through its financial and technical support. He expresses profound gratitude to his supervisor, Dr Bright Yeboah-Akowuah (PhD), and his mentor, Dr Henry Nunoo (PhD), and to the KEEP management, led by Prof. Jerry John Kponyo, for their continuous encouragement.