Inferred Check

Qualytics automatically generates and maintains inferred checks by Profiling the Datastore, performing statistical analysis followed by various machine learning methods.

Info

Inferred checks will be automatically updated with the next Profiling run.
Manually updating an inferred check will take it out of the automatic update workflow.

Inference Engine

1. After metadata is generated by a Profile Operation, Inference Engine is initiated to kick off Inductive and Unsupervised learning methods

2. Available data is partitioned into a training set and a testing set.

3. The engine applies numerous machine learning models & techniques to the training data in an effort to discover well-fitting data quality constraints.

4. Those inferred constraints are then filtered by testing them against the held out testing set & only those that assert true above a certain threshold are converted and exposed to users as Inferred Checks.