BS6206 Experiential Data Science
Summary of course content
In this specialization-track experiential learning course, you will apply existing knowledge on statistics and machine learning in an industry setting. You will be placed in real-life client projects to get hands-on experience to apply theory to practice.
You will work with practicing management consultants and data translators. You will be involved in the process of delivering and creating impact for clients from advanced analytics models.
Aims and objectives
1. You will learn to apply data science theories in a competitive real-world setting
2. You will learn about the various career paths available to data scientist to pursue
3. Via experiential learning in teams (4-5), you will acquire the necessary soft skills (e.g. teamwork, communication, leadership and grit) necessary for successful outcomes
Syllabus
1. Computational thinking and its applications in data science
2. Essential logic for data science
3. The winner’s curse (p-value instability issues)
4. The Anna Karenina Principle (when statistical significance is actually misleading)
5. Dealing with batch effects in small and big data
6. Weak validation practices in machine learning and AI
7. How to assess a prediction model (for reproducibility, explainability and meaning)
8. Special topics in industry (guest lectures from industry partner)
Assessment
Individual progress journal (reflection and technical organization) | Individual | 40% |
First draft model and results | Group | 10% |
1st Draft presentation | Group | 15% |
2nd presentation | Group | 15% |
Final presentation | Group | 20% |
100% |