5.1 Customer AI - Data Preparation (Ingest)
Customer AI - Data Preparation (Ingest)
Last updated
Customer AI - Data Preparation (Ingest)
Last updated
In order for Intelligent Services to discover insights from your marketing events data, the data must be semantically enriched and maintained in a standard structure. Intelligent Services leverages Adobe's Experience Data Model (XDM) schemas in order to achieve this. Specifically, all datasets that are used in Intelligent Services must conform to the Consumer Experience Event XDM schema.
In this exercise, you'll create a schema that contains the Consumer Experience Event mixin, which is required by the Customer AI Intelligent Service.
Log in to Adobe Experience Platform by going to this URL: https://experience.adobe.com/platform.
After logging in, you'll land on the homepage of Adobe Experience Platform.
Before you continue, you need to select a sandbox. The sandbox to select is named --module10sandbox--. You can do this by clicking the text [!UICONTROL Production Prod] in the blue line on top of your screen. After selecting the appropriate sandbox, you'll see the screen change and now you're in your dedicated sandbox.
From the left menu, click Schemas and go to Browse. Click Create Schema.
In the popup, select XDM ExperienceEvent.
You'll then see this.
Search and select the following Mixins to add to this Schema:
Consumer Experience Event
End User ID Details
Click Add Field Groups.
You'll then see this. Select the Mixin End User ID Details.
Navigate to the field endUserIDs._experience.emailid.id.
In the right menu for the field endUserIDs._experience.emailid.id, scroll down and check the checkbox for Identity, check the checkbox for Primary Identity and select the Identity namespace of Email.
Navigate to the field endUserIDs._experience.mcid.id. Check the checkbox for Identity and select the Identity namespace of ECID. Click Apply.
Give your schema a name now.
As the name for our schema, you'll use this:
--demoProfileLdap-- - Demo System - Customer Experience Event
As an example, for ldap vangeluw, this should be the name of the schema:
vangeluw - Demo System - Customer Experience Event
That should give you something like this. Click the + Add button to add new Mixins.
Select the name of your schema. You should now enable your schema for Profile, by clicking the Profile toggle.
You'll then see this. Click Enable.
You should now have this. Click Save to save your schema.
From the left menu, click Datasets and go to Browse. Click Create dataset.
Click Create dataset from schema.
In the next screen, select the dataset you created in the previous exercise, which is named [!UICONTROL ldap - Demo System - Customer Experience Event]. Click Next.
As a name for your dataset, use --demoProfileLdap-- - Demo System - Customer Experience Event Dataset. Click Finish.
Your dataset is now created. Enable the Profile toggle.
Click Enable.
You should now have this:
You're now ready to start ingesting Consumer Experience Event data and start using the Customer AI service.
Once the Schema and Dataset are configured, you're now ready to ingest Experience Event data. Since Customer AI requires data across 2 quarters at least, you'll need to ingest externally prepared data.
The data prepared for the experience events must comply to the requirements and schema of the Consumer Experience Event XDM Mixin.
Please download the file containing sample data from this location: https://dashboard.adobedemo.com/data. Click the Download button.
Alternatively, if you can't access the above link, you can download the file also from this location: https://aepmodule10.s3-us-west-2.amazonaws.com/retail-v1-dec2020-xl.json.zip.
You've now downloaded a file named retail-v1-dec2020-xl.json.zip. Place the file on your computer's desktop and unzip it, after which you'll see a file named retail-v1.json. You'll need this file in the next exercise.
In Adobe Experience Platform, go to Datasets and open your dataset, which is named ldap - Demo System - Customer Experience Event Dataset.
In your dataset, click Choose files to add data.
In the popup, select the file retail-v1.json and click Open.
You'll then see the data being imported, and a new batch is created in the Loading state. Don't navigate away from this page until the file is uploaded.
Once the file has been uploaded, you'll see the batch status change from Loading to Processing.
Ingesting and processing the data might take 10-20min.
Once data ingestion is successful, the batch status will change to Success.
