We use cookies. You have options. Cookies help us keep the site running smoothly and inform some of our advertising, but if you’d like to make adjustments, you can visit our Cookie Notice page for more information.
We’d like to use cookies on your device. Cookies help us keep the site running smoothly and inform some of our advertising, but how we use them is entirely up to you. Accept our recommended settings or customise them to your wishes.
×

What to know about Adobe Experience Platform’s data science capabilities

Welcome back to the Adobe Experience Platform (AEP) education series. In our last blog post we took a look at How Adobe Experience Platform Drives Data Transformation and discussed how data transformation can drive your connection with customers. In this installment, we’ll take a closer look at how the data science capability within AEP (called Data Science Workspace) can help brands evolve towards real-time machine learning (ML).

AEP has been helping us to centralize and standardize our customer data and content across the enterprise – powering 360° customer profiles and enabling data governance to drive personalized experiences.

Further, to enhance the experience and build more intelligent services, AEP now provides services that not only include capabilities for data ingestion, but also analyzing data, building predictive models, and identifying next-best-actions leveraging data science as an integral part of the platform. This will now enable the platform to make the data, insights, and decisions available to experience-delivery systems to act upon in real time, yielding compelling experiences in the relevant moment.

With data science integrated into AEP, enterprises will be able to utilize completely coordinated data-driven marketing and analytics solutions for driving meaningful customer interactions, driving positive business results.

What can data science unleash?

Organizations, today, put a high priority on mining big data for predictive analytics and insights that will help them personalize customer experiences and deliver more value to customers— and to the business.

Data science uses machine learning and artificial intelligence to unleash insights from your data. Integrated within AEP, data science enables customers to create predictive models utilizing data across AEP and Adobe Solutions to generate intelligent insights and decisions to weave delightful end-user digital experiences.

With data science, data scientists can easily create intelligent services that are APIs powered by machine learning. These services work with other Adobe services, including Adobe Target and Adobe Analytics Cloud, to help further automate personalized, targeted digital experiences in web, desktop, and mobile apps.

Data science as a workspace

With Data Science Workspace, AEP allows us to bring experience, focused AI to bear across the enterprise, streamlining and accelerating data-to-insights-to-code with the following:

  • A Machine Learning (ML) framework and runtime
  • Integrated access to our data stored in AEP
  • A unified data schema built on the Experience Data Model (XDM)
  • The compute power essential for machine learning/AI and managing big datasets

Data scientists of all skill levels will get sophisticated, easy-to-use tools that support rapid development, training, and tuning of machine learning models.

With Data Science Workspace, data scientists can streamline the cumbersome process of uncovering insights in large datasets. Data Science Workspace delivers advanced workflow management, model management, and scalability.

Some of the features that make Data Science Workspace very constructive are:

1. One-stop data access to explore and prepare data

Data Science Workspace is fully integrated with AEP, including the data lake, unified profile, and unified edge. This helps explore all organizational data stored in AEP at once, along with common big data and deep learning libraries, such as Spark ML, and TensorFlow. It also supports options to ingest your own datasets using the XDM standardized schema.

2. Jupyter-Lab on AEP

Jupyter-Lab is tightly integrated into AEP. It provides an interactive development environment for data scientists with architectural changes, design considerations, customized notebook extensions, and pre-installed libraries to work with Jupyter notebooks, code, and data. Jupyter-Lab on the AEP also provides support to major industry standard languages, like Python, R, Py-Spark, and Spark (Scala).

3. Pre-built machine learning recipes, authoring, and experimenting

AEP data science

A recipe is a top-level container representing a specific machine learning, AI algorithm, by processing logic, and configuration required to build and execute a trained model and hence help solve specific business problem in significantly shorter time.

Data Science Workspace includes prebuilt ML recipes for common business needs so users don’t have to start from scratch. Workspace also provides support to either modify a prebuilt recipe, or build one from scratch, by authoring recipes in Jupyter Notebook. Once ready, we can create one or more unique instances of the recipe and change the parameters for each instance to experiment, then test each unique recipe instance’s performance.

Once tested, finalized recipes can be used for audience scoring. After the scoring run is completed, a calculated response can be written to the AEP dataset. Further, this data set can be fed to AEP Real Time Profile Service to create segments within AEP.  This scoring run data is also available to be used within the segmentation builder of the AEP Segmentation Service if the dataset is marked for profile.

Data science applied in real time

Real-time ML can dramatically enhance the relevance of your digital experience content for your end-users. This is made possible by leveraging real-time inferencing and continuous learning from the experience data.

Organizations can evolve from batch predictive models that are weeks to a month old, to real time. As an example, this becomes even more important to score an anonymous prospect’s behavior from current or past sessions and to provide the optimal offer to increase conversions.

Customer behaviors like search for coupons or request to return can be considered to provide a discounted offer. Models like churn or propensity can now be run in real time with the latest data to provide increase conversion or reduce churn.

In conclusion, the data science capability in AEP can help empower brands to gain and apply insights by optimizing customer experiences.

If you’re interested in learning more, stay tuned for our next blog post: How AEP and Merkury Identity can drive real time experiences.

Adobe Summit ad

Join the Discussion