02 - SCSM Email Work Item Classification (Create Web Service w/ Trained Models)
Driven by the SMLets Exchange Connector, this experiment will define Incident/Service Request, Classification, and Support Group for New Work Items in System Center Service Manager.
This experiment (2 of 2) is taken directly from the following [Machine Learning Gallery Project](https://gallery.azure.ai/Experiment/E-mail-Classification-for-Automated-Support-Ticket-Generation-Step-2-of-2-Create-Web-Service-with-Trained-Models-2) leveraging a sample, but modified dataset from [Microsoft/Endava seen here on GitHub](https://github.com/karolzak/support-tickets-classification) in order to build a machine learning classification engine that can be consumed as a Rest API for the SMLets Exchange Connector. The [SMLets Exchange Connector](https://github.com/AdhocAdam/smletsexchangeconnector) is an open source PowerShell based alternative to the stock Exchange Connector provided by Microsoft for System Center Service Manager that is used to parse an Exchange inbox and create or update Work Items in SCSM. To configure these AML experiments for your Service Manager deployment, a full walkthrough on the project's Github can be found [here](https://github.com/AdhocAdam/smletsexchangeconnector/wiki/Azure-Machine-Learning-setup).