Analytics
News, events, thought leadership and more.
Welcome back to our Alteryx Analytics Cloud Platform monthly blog, where we’ll be sharing some exciting new product capabilities introduced in January 2023! As you might have noticed, last week we announced new self-service and enterprise-grade capabilities to its Alteryx Analytics Cloud Platform to help customers make faster and more intelligent decisions across the enterprise. Please stay tuned for more information, which we’ll be sharing in our next February blog.
Alteryx Machine Learning
- Partial Dependence Performance Enhancement
- Feature Engineering Improvements
- Anomaly Alerts on Predictions
Alteryx Auto Insights
- Common Causes
Keep reading to learn more about these exciting new capabilities and features.
Alteryx Machine Learning
We’ve been listening to feedback from our customers and are delivering multiple new performance enhancements and feature improvements for Machine Learning in January!
First, we have a key performance enhancement in the model evaluation step. Partial Dependence plots show you the effect that a single predictor has on your model’s outcome. Users will notice the processing speed for this step has been greatly increased, making model evaluation faster and easier than ever.
Partial Dependence Plots show the marginal impact of predictors on model outcomes
Machine Learning makes Feature Engineering fast and easy. With this release, we’ve made this process even better. Now users can manually select which features they want to go into models, enabling greater flexibility in framing a model around a specific use case and maximizing the predictive power of the model.
This gif shows a user eliminating similar features from a housing model
Finally, we’ve added a new alert that lets users know when something is amiss with their data. Sometimes customers will load a new prediction data set into their model that has key differences from that data used to train the model. This new alert will automatically flag the differences in the data, saving users from faulty predictions.
Anomaly Detection within AML identified differences between the prediction dataset and original training dataset.
Alteryx Auto Insights
Common Causes is here! Common causes analyzes two different metrics in your data and provides insights into the relationship between the two metrics. Users can quickly understand if two metrics move in the same direction, the strength of correlation between them, and more. This allows business users to gain an even deeper understanding of what factors are impacting their most critical KPIs by providing them with an explanation of how performance is connected across two metrics over time. Read about Common Causes in more detail here.
Conclusion & Learn More
Thank you for reading and if you want to learn more about this release, check out Release Notes section below. And don’t forget to let us know your thoughts or share your ideas on our Community Discussion Forums. See you next month!
Release Notes
Alteryx Designer Cloud | Release Notes
Alteryx Machine Learning | Release Notes
