Find the Problem
A real, deliberately flawed dataset. Click the rows you think have a problem, then check your answers.
Demonstration using representative data: not real product data.
What is happening?
Click any row you think has a data quality problem, then check your answers. A row is flagged as a genuine problem if it has a missing value, a negative value in a numeric column, or a value that's a statistical outlier (IQR rule) within its own column: the same checks the Data Quality Checker runs, applied per row instead of aggregated into a score.
Under the Hood
CSV parsed in-browser↓Your flags compared against computed issues per row↓Score: correct catches, false flags, missed issues
Inputs
- A 14-row product dataset
Processing
- Missing/negative/IQR-outlier detection per row
Output
- Per-row verdict with explanation
Technology
- Vanilla JavaScript
Explore another experiment
More Playground experiments are on the way.