Considerations and Limitations for Self Learning Knowledge

Considerations for Self Learning Knowledge:

• When configuring Customers Signals Intelligence, schedule your batch data transforms monthly. This enables Self Learning Knowledge to get the most recent insights.
• Self Learning Knowledge dashboard works on a "point in time" principle. Every time data processing is run, new contact reasons are generated. Some of the newly generated contact reasons could be duplicative with previously generated contact reasons, because those historical contact reasons were true at that time. This is intended behavior.

Known Issues and Limitations for Self Learning Knowledge:

• The setup of Self Learning Knowledge, specifically the data transform portion, can take up to 24 hours to complete on the initial run, based upon the volume of data processed.
• Currently, generating a knowledge article from a suggestion won’t delete or archive that suggestion. Users track the suggestions that they’ve acted on manually. This is a known issue that will be fixed in a later release.
• Currently, Self Learning Knowledge is not supported in companion orgs. This is a known issue that will be fixed in a later release.
• Customer Signals Intelligence deletes prior records in the Dedup EATI associations table every time the clustering run happens. As a result, old suggestions are no longer actionable even though they’re present in the Tableau dashboard. If you run the clustering service too frequently, it results in suggestions for knowledge articles you can’t create. This is a known issue that will be fixed in a later release.