pii

Presidio leverages ML models which might detect an instance in one sentence but not in another.
By automatically adding all instances of a previously identified entity, we can increase detection recall (and potentially decrease precision)

Example (hypothetical):

Text: "TrueForce is an American company. Lately, TrueForce was acquired by FalseForce".

If the first "TrueForce" was detected

It is not surprising that deep and shallow scan show different results. Shallow scan only looks at column names. Deep scan looks at a sample of the data. I've even noticed that two different runs of deep scan show different results as sample rows are different. This is the challenge with not scanning all of the data. Its a trade-off between performance/cost and accuracy. There is no right answer.