Enrich Real-Time Customer Profile with machine learning insights

Adobe Experience Platform Data Science Workspace provides the tools and resources to create, evaluate, and utilize machine learning models to generate data predictions and insights. When machine learning insights are ingested into a Profile-enabled dataset, that same data is also ingested as Profile records which can then be segmented using Adobe Experience Platform Segmentation Service.

The segmentation process depends on the evaluation method for the audience. If an audience is configured as streaming, it will process all the new updates written by the model into the profile in real-time. However, if an audience is configured for batch evaluation, the new values will be evaluated in the next batch.

This document provides links to tutorials that enable you to enrich Real-Time Customer Profile with your machine learning insights.

Getting started

In order to complete the tutorials below, you are required to have a working understanding of ingesting Profile data and creating audiences. Before beginning this tutorial, please review the documentation for the following services:

In addition to the above-mentioned documents, it is highly recommended that you also review the following guides on schemas and the Schema Editor:

Create and configure an output schema and dataset create-an-output-schema-and-dataset

The first step towards enriching Real-Time Customer Profile with scoring insights is knowing what real-world object (such as a person) your data defines. Having an understanding of your data enables you to describe and design a structure to add meaning, much like designing a relational database.

Composing a schema begins by assigning a class. Classes define the behavioral aspects of the data the schema will contain (record or time-series). To start making your own schemas, follow the steps in the tutorial on creating a schema using the Schema Editor. Note that before you can enable a dataset for Profile, you need to configure the dataset’s schema to have a primary identity field and then enable the schema for Profile. When data is ingested into a Profile-enabled dataset, that same data is also ingested as Profile records.

If you prefer to compose a schema using the Schema Registry API instead, start by reading the Schema Registry developer guide before attempting the tutorial on creating a schema using the API.

Once your schema and dataset are prepared, you can generate and ingest scoring data to the dataset by performing scoring runs using an appropriate model.

Create audiences using the Segment Builder create-audiences-using-the-segment-builder

After you have generated and ingested your scoring data insights to your Profile-enabled dataset, you can create dynamic audiences using the Segment Builder.

The Segment Builder provides a rich workspace that allows you to interact with Profile data elements. The workspace provides intuitive controls for building and editing rules, such as drag-and-drop tiles used to represent data properties. Follow the Segment Builder user guide to learn about:

Next steps next-steps

To learn more about audiences and the Segment Builder, read the Segmentation Service overview.

To learn more about Real-Time Customer Profile, read the Real-Time Customer Profile overview