Tutorial – Using Qlik AutoML to create a prediction app
This tutorial teaches you how to use Qlik AutoML to analyze data and create apps to visualize prediction data created with the platform.
We will consider the scenario of customer churn, a classic example of a binary classification problem. The goal is to be able to reliably predict whether a customer will cancel their subscription or remain a subscriber to a service. There are only two outcomes in this type of problem: true or false (churned, or not churned).
To approach this machine learning problem, we will start by processing a set of data for which we already know the outcome, then apply statistical modeling created from that data to new data we would like to predict the outcomes for. This is the approach we will take in this tutorial.
To create reliable and accurate models, you should ensure that your training dataset does not contain leakage or "leaky" features. Data leakage occurs when one or more features in the training data can be used to derive the target variable you are trying to predict, or when one or more features in the training data includes information that would not be known at the time of prediction.
You will begin this tutorial by creating an experiment. From there, you will refine and deploy the experiment into a machine learning model. This model will be used to create predictions, which can be shown in the form of visualizations in a Qlik Sense app.
What you will learn
Once you have completed this tutorial, you will understand the different steps involved in creating and configuring an experiment. You will also learn how to interpret model scores. Finally, you will be able to deploy a machine learning model and will understand how your predictions data can be used to generate compelling Qlik Sense visualizations in Qlik Cloud Analytics.
Who should complete this tutorial
This tutorial is designed for users who want an introduction to automated machine learning in Qlik Cloud Analytics.
What you need to do before you start
Download this package and unzip it on your desktop:
The package contains the two data files needed to complete this tutorial. Upload the data files to the catalog.
The training dataset contains information about customers whose deadline for renewal has passed, and have made the decision to churn or remain subscribed to the service.
The apply dataset contains details about a new set of customers whose renewal date has not yet passed. It has not yet been determined whether or not these customers will cancel their service. The goal, with this tutorial, is to predict what this set of customers will do, with the hope that we can decrease the likelihood that they will churn.
Do the following:
-
Open the Qlik Cloud Analytics hub.
-
Click Add new > Dataset, and then select Upload data file.
-
Drag the Customer churn data - training.csv file to the upload dialog.
-
Next, drag the Customer churn data - apply.csv file to the upload dialog.
-
Select a space. It can be your personal space or a shared space if you want other users to be able to access this data.
-
Click Upload.
Now that your datasets are uploaded, you can proceed to creating an experiment.
Lessons in this tutorial
The topics in this tutorial are designed to be completed in sequence. However, you can step away and return at any time.
Further reading and resources
- Qlik offers a wide variety of resources when you want to learn more.
- Qlik online help is available.
- Training, including free online courses, is available in the Qlik Continuous Classroom.
- Discussion forums, blogs, and more can be found in Qlik Community.
Your opinion matters
We appreciate any feedback that you may have. Please use the section below to let us know how we're doing.