Product External IDs in Ticket NLP Classification

[valentina190285]

Dear Members,
in our Service Ticket scenario, in the e-mail received, are mentioned Product External IDs (ERP Material IDs) and not C4C Product IDs (and those IDs are not the same).

Ticket NLP Classification extracts a product ID number from the text of the email received and fills in the Product and Product ID fields, but (unfortunately) doesn't consider Product External IDs.

As in many cases SAP ERP Product IDs are well known also by customers and those IDs are mentioned in service ticket e-mails, this functionality could allow users to save time in ticket compilation.

Is there any possibility to reach this behaviour (reading external SAP Product IDs) using sdk?

Thank you,
Regards,
Valentina

[former_member226]

Hi,

To my knowledge, there is no way that you can influence the standard model to predict Ext Product ID.

However, you can use BADI MachineLearningInputDataPreProcessing of namespace http://sap.com/xi/AP/CRM/MachineLearning to influence/adpat the text even before it hits the ML model for prediction. Here the approach could be that read the input text to replace the Product Ext ID(if exists in text) with Actual C4C Product ID(using objectIdentifierMapping BO) and then ML model can do the rest, as usual. An example implementation could be found HERE.