About Script Score

The score of a prediction produced by a machine learning or fuzzy logic script represents the goodness of fit, in the statistical sense, for the specific prediction. When testing a script, scores are presented as a percentage between 0.0% and 100.0%. When passing the script score to a successor node in a Cognitive Analytics cognition, scores are presented as a numeric value between 0 and 1. Higher scores mean that the prediction, according to the script's logic, is more likely to be a correct prediction.

Example

The failureMechanism.py script is a fuzzy logic script used to identify the failure mechanism involved in a specific equipment failure.

The following table provides the text from sample work history events and the failureMechanism.py script's predictions for them and the corresponding scores.

Event Short Description Text	failureMechanism.py Prediction	failureMechanism.py Prediction Score
REPLACE PUMP BEARINGS; OIL ANALYSIS INDICATES LARGE WEAR PARTICLES DENOTING SEVERE SLIDING AND BEARING WEAR	Wear	100.0%
INSPECT IMPELLER FOR FOREIGN MATERIAL IT WAS PLUGGED OFF.. TORN DOWN, CLEANED AND REBUILT PUMP.. MOTOR WAS ALSO SENT OUT AND REBEARING	Blockage/Plugged	85.0%
REPLACE IMPELLER; CHANGE TO AN ELEVEN INCH IMPELLER	Wear	61.2%
HAS BAD SEAL. CHECK WITH OPER	Leakage	59.7%