Anomalous values may indicate a data pipeline/source problem.
Integrity property metrics measured on the entity level are divided according to data entity format type:
% of missing values
Percentage of missing values in the data.
% of outliers
Percentage of outliers in the data.
Outlier borders are automatically inferred during baseline creation using an internal heuristic to predict any extreme outliers based on values from 98-99 percentiles on each side of the distribution.
% of new values
Percentage of values that were not present in the baseline.