COMP 2309 · Data Science: Business Integration

Led by Vapnikian Statistical Learning Simulacrum

SQL, Python, and Tableau integrated in a real business case — the complete absenteeism prediction pipeline from raw data to Tableau dashboard.

1. Module 1

   SQL, Python, and Tableau: The Integration

   Led by Vapnikian Statistical Learning Simulacrum

   The question

   SQL for querying, Python for modelling, Tableau for communicating. What does each tool contribute that the others cannot — and what is the correct sequence in the absenteeism pipeline?

   Outcome

   The student can describe the three-layer ecosystem and when each tool is appropriate.

   Sub-units

   1. ○ 1.1 The Three-Layer Pipeline Open this unit → ✓ ↻
   Begin the course → Print certificate of completion
2. Module 2

   The Absenteeism Business Case: Preprocessing

   Led by Vapnikian Statistical Learning Simulacrum

   The question

   28 reason codes, a date column, continuous and categorical features mixed together. What are the preprocessing decisions — and how do you create a binary target from a continuous outcome?

   Outcome

   The student can execute the full absenteeism preprocessing pipeline.

   Sub-units

   1. ○ 2.1 Preprocess the Data Open this unit → ✓ ↻
   Open this module → Print certificate of completion
3. Module 3

   The Absenteeism Business Case: Modelling

   Led by Vapnikian Statistical Learning Simulacrum

   The question

   Fit a logistic regression, interpret the coefficient table, remove near-zero predictors. Which features most strongly predict excessive absence — and what does the model say about the role of age, reason, and commuting cost?

   Outcome

   The student can build, evaluate, and interpret a logistic regression for a business case.

   Sub-units

   1. ○ 3.1 Build and Interpret Open this unit → ✓ ↻
   Open this module → Print certificate of completion
4. Module 4

   Visualisation and Communication: Tableau

   Led by Vapnikian Statistical Learning Simulacrum

   The question

   A manager who cannot read a regression table will act on a scatter plot with a clear message. How do you translate a logistic regression coefficient into a Tableau visualisation that drives a HR decision?

   Outcome

   The student can produce three business-interpretable visualisations from model predictions.

   Sub-units

   1. ○ 4.1 Three Visualisations Open this unit → ✓ ↻
   Open this module → Print certificate of completion
5. Module 5

   ChatGPT and AI Tools in the Data Science Workflow

   Led by Vapnikian Statistical Learning Simulacrum

   The question

   ChatGPT can write preprocessing code and generate EDA suggestions. Does this raise or lower the bar for what a data scientist needs to know? You now evaluate AI-generated code rather than write it — is that easier or harder?

   Outcome

   The student can use AI tools productively and evaluate their outputs critically.

   Sub-units

   1. ○ 5.1 Final Essay: The AI-Augmented Data Scientist Open this unit → ✓ ↻
   Open this module → Print certificate of completion