A false positive is an outcome through an application of a functionality or set of functionalities in IDEA which wrongly indicates that a particular condition or attribute is present in the data set.
False positives can be extremely disruptive in the analytic process and frustrate the end goal.
It is important to identify false positives across various business processes and their check points while you build the analytics.
As a popular and common example - running a simple duplicate key detection on vendor code and vendor invoice number in a vendor ledger would certainly result into a list of potential duplicates.
A closer examination of the potential duplicates will reveal that the displayed duplicates could have been reversed at some time during the booking process.
Not giving credence to such invoice reversals is a false positive.
So by using IDEA's summarization by invoice booking document number you can negate reversals and take only unique invoices into account for real duplicate checks.
Elimination of false positives is a patient process which spans the analytic maturity curve. Knowledge of false positives and elimination will grow with a more detailed understanding of the process and its control points.
Just as a thought - a lot of emphasis is placed in the analytics community for auditors on creating a repository of control points. May be its also time to create a repository of commonly found false positives against each of these control points.