Qlik AutoML on Qlik Cloud - sample data

[QlikProductUpdates]

In coordination with the launch of Qlik AutoML on Qlik Cloud, we are making a set of sample data files available for people to use when trying out the new capability. These files represent a fictitious company with customer subscriptions, looking to predict which customers will stay and leave in the future. This simple data will allow our users to quickly experience the power of machine learning as they learn the AutoML process.

See attached files at the end of this post

Step 1 - When creating your first AutoML Experiment, use the Customer_Cancellations_Training.csv dataset to train your model. This dataset contains historical customer churn data for the machine to learn from. 

Step 2 - After you have deployed your first model, you can use the Customer_Cancellations_Apply.csv dataset to create predictions. This dataset contains current customer data on which churn predictions can be made.

Watch this video to get an understand on how to use this data.

Can't see the video? Watch on the Qlik video site.

[QlikProductUpdates]

AutoML on Cloud now GA!  Read more here:  https://community.qlik.com/t5/Qlik-Product-Innovation-Blog/Introducing-Qlik-AutoML-on-Qlik-Cloud/ba-...

[marcel_olmo]

Nice video! But could you add your finished Customer Retention App in Qlik Sense as well? It would be great!

[Roeeh]

hey , it is not clear what is the difference between the 2 files 

what is currant data and historical data ?  

[Veerendra_Katikala]

Yes...it is not clear me too what is Training and Apply files , could you please elaborate more?

[Kyle_Jourdan]

"Training" data refers to the data used to initially train the machine learning model. "Apply" dataset is a separate set of data that was not used in model training that we are applying predictions to.

As a real-world example, we may use historical customer information to train a customer churn model; this historical data is our "training" dataset. Once our model is deployed, we would "apply" these predictions to a list of our current customers; this is current customer data is our "apply" dataset.