bug
The backend's absolute outlier math flags valid real-world data trends (like long-term stock market growth or fraud target rows) as data validation errors, completely painting columns in bright red. Because there is currently no way to tell the app "this column is supposed to look like this," users get stuck with permanent false-alarm highlights that ruin table scannability.
Possible Fix?
Frontend fix would be to add a simple "Mute Anomaly Highlights" checkbox or toggle switch at the top of each sidebar column card. Clicking it would immediately clear the red background fills from that specific column in the data table below, allowing analysts to silence false alarms manually --> just an idea, i do not know if it's possible as a "quick fix".
Dataset used
- This happened with a lot of financial data!
MSFT_Stock_Data_1986_2026.csv
synthetic_fraud_dataset.csv
Screenshot
bug
The backend's absolute outlier math flags valid real-world data trends (like long-term stock market growth or fraud target rows) as data validation errors, completely painting columns in bright red. Because there is currently no way to tell the app "this column is supposed to look like this," users get stuck with permanent false-alarm highlights that ruin table scannability.
Possible Fix?
Frontend fix would be to add a simple "Mute Anomaly Highlights" checkbox or toggle switch at the top of each sidebar column card. Clicking it would immediately clear the red background fills from that specific column in the data table below, allowing analysts to silence false alarms manually --> just an idea, i do not know if it's possible as a "quick fix".
Dataset used
MSFT_Stock_Data_1986_2026.csv
synthetic_fraud_dataset.csv
Screenshot