Start the Data Ingestion and Model Training

Now, you can build a prediction model based on data pulled from closed and resolved requests created from inbound emails.

Before you start

Data from the last 6 months is used for the model training so the latest data is always included. You have the option to retrain the model on the latest data every 2 weeks. Every time the model is retrained, older data is discarded and new data is added.

Here's what to do

1. Sign in to Fusion Service with either the CRM SVC Attribute Detection Administrator or Adaptive Intelligence Applications Administrator role.
2. Click the Tools menu.
3. In the Quick Actions pane, click AI Apps Administration.
4. From the list of AI applications on the Application Administration page, click Service Request Classification.
   The Configure SR Classification page is displayed, and the Configuration tab is selected by default.
5. In the Credentials to connect to Fusion Service area, specify the following details:
   1. In the User Name field, enter the user name for your Fusion Service environment.
   2. In the Password field, enter the password for your Fusion Service environment.
   3. In the Endpoint field, enter the Oracle Fusion Service BIP Host name.
   4. Click Save.
      Note: After the credentials are saved, you see a Credentials Saved message, and the Activate button in the Activate Tenant area is enabled. This indicates that the AI application can ingest the data from Fusion Service.
   5. To ignore the changes, click Cancel.
6. In the Activate Tenant area, do the following:
   1. Click the Model Retraining (Biweekly) check box if you want the model to be retrained on the latest data every 2 weeks.
   2. Click Activate to schedule the data ingestion and model training. If it's been activated before, click Reactivate.

Results:

The data ingestion and model training begins and the process runs in the background.