TabularDates unconfirmedPump it Up: Data Mining the Water Table
Can you predict which water pumps are faulty to promote access to clean, potable water across Tanzania? This is an intermediate-level practice competition.
Official page previewCaptured Oct 19 · not live
Open the competition on DrivenData↗
Open the competition on DrivenData↗- Effort
- 5–15 hours
- Difficulty
- Intermediate
- Compute
- CPU only
- Scoring
- Not verified
Is it for you?
Prepare before joining
Beginner fit60/100
Learning value88/100
Who it suits
Learners comfortable with basic ML who want to practice multiclass classification, geographic feature engineering, and handling messy real-world data with high-cardinality categoricals.
What makes it hard
The dataset has many high-cardinality categorical features, geographic coordinates that benefit from spatial feature engineering, and a three-class imbalanced target that requires careful encoding and model tuning.
Getting started
Skills to bring
- Python
- pandas and data wrangling
- scikit-learn or gradient boosting libraries
- feature engineering techniques
From the organizer
- DataWater-pump dataset with operational, installation, and management variables.
First competition? Read No GPU? Pick Competitions That Still Teach You Good Habits