Predictive analytics

Introduction to Data Literacy

Anneleen Rummens

Data Scientist

Analytics overview

Be careful: there is always a degree of uncertainty associated with predictions

Zoom-in of predictive analytics

Machine learning models
- Classification-based
  - Predicting cancellations of subscriptions
- Regression-based
  - Predicting housing prices based on neighborhood characteristics
Time series forecasting
- Predicting sales revenue over time
Predictive text analysis
- Predicting whether an email is spam or not

Example time series

Steps of predictive modeling

Data is split into training and test set for building the predictive model
Predictions are interpreted and evaluated on the test data, using pre-determined metrics like accuracy (percentage of correct predictions)

Which team is most likely to win the next FIFA World Cup?

Select relevant variables like team ratings, player ratings, rankings, and match difficulty
Build a predictive model to predict probability of winning or reaching specific phases

Use insights to predict winning probability for each country

World Cup trophy and ball

Introduction to Data Literacy