Mosa Nyamande
Profile Summary
Mosa is passionate about developing Africa through technology. Mosa is a Co-founder of Khonology, a company with a mission to Digitise Africa. Mosa has led the implementation of multiple world class FinTech solutions in the past decade, ranging from client onboarding, to data analytics, to digital payments to regulatory compliance for Africa’s leading financial services and insurance institutions.
​
Mosa’s expertise covers Data Science, Advanced Analytics, Data Architecture and Digital Strategy. These competencies unlock value for any organisation. Mosa is a Microsoft Certified: Azure Data Engineer Associate and AWS Certified Cloud Practitioner.
​
Mosa’s current role is as Director of Delivery, where he looks after Khonology’s Delivery System – which is Khonology’s special way of delivering Digital Solutions.
​
Mosa achieved an Honours degree from Wits University and started in entrepreneurship during those days, where he started another business before Khonology with the same co-founders — The Consulting Academy — A strategy consulting company run by students.
​
Mosa loves jazz and he has played the drums for the last 14 years. Mosa also loves being active, and spends a lot of time outdoors and at the gym.
Session Description
Practical Machine Learning Modelling.
Topics & Durations (Total 1.5 hours):
Intro to Advanced Analytics – 5 minutes
-
Converts hindsight into foresight: a comparison of descriptive, diagnostic, predictive and prescriptive analytics
-
The Advanced Analytics maturity path
Intro to Machine Learning – 20 minutes
-
Machine Learning vs Traditional Programming
-
Machine Learning Explained
-
Feature Engineering to prepare for Machine Learning
-
Machine Learning Applied
-
Model Testing & Validation
-
Model Reinforcement & Retraining
-
Summary: 7-Step Machine Learning Methodology
-
Algorithm Selection Guidance
Practice Exercise: Supervised Machine Learning, Classifier Model – 1 hour
-
Apply the 7-Step Machine Learning Methodology:
-
​Get the data
-
Pre-process the data
-
Choose an Algorithm to Train the Machine
-
Apply the Algorithm to Training Data
-
Apply the Algorithm to Test Data
-
Evaluate Model fit on Test Data
-
Revise the Model to generalise better on out of sample data
