Lab: Improve the Accuracy of Anomaly Detection Models

Lab: Improve the Accuracy of Anomaly Detection Models

Use Case: Predictive Maintenance

About this Mission

This mission builds on the Detect Times Series Anomalies lab. On this mission, you will apply strategies for improving the accuracy of your anomaly detection models by leveraging Feature Derivation Windows, repository blueprints, additional feature lists, advanced tuning, and blenders.

Mission format and duration: self-paced, hands-on, 45 minutes

Upon completion of this mission, you will be able to: 

  • Select an appropriate FDW for a dataset
  • Run blueprints on the predictive maintenance dataset and on alternate feature lists
  • Examine models using the Anomaly Assessment tool
  • Perform advanced tuning and create blenders to try to improve the accuracy of an anomaly detection model
  • Identify leading indicators of anomalies in your datasets

Who should complete this mission?

  • Business Analysts
  • Citizen Data Scientists
  • Data Scientists

Before embarking on this mission, you should complete one of the following:

We also recommend that you also complete the following before beginning this mission:

Technical requirements

  • Chrome browser
  • DataRobot Automated Machine Learning — If you don’t have access to the application, please sign up for our free trial: datarobot.com/trial.



Action Items45 min

  • Complete the Lab
  • Resources
  • Datasets

About this Mission

This mission builds on the Detect Times Series Anomalies lab. On this mission, you will apply strategies for improving the accuracy of your anomaly detection models by leveraging Feature Derivation Windows, repository blueprints, additional feature lists, advanced tuning, and blenders.

Mission format and duration: self-paced, hands-on, 45 minutes

Upon completion of this mission, you will be able to: 

  • Select an appropriate FDW for a dataset
  • Run blueprints on the predictive maintenance dataset and on alternate feature lists
  • Examine models using the Anomaly Assessment tool
  • Perform advanced tuning and create blenders to try to improve the accuracy of an anomaly detection model
  • Identify leading indicators of anomalies in your datasets

Who should complete this mission?

  • Business Analysts
  • Citizen Data Scientists
  • Data Scientists

Before embarking on this mission, you should complete one of the following:

We also recommend that you also complete the following before beginning this mission:

Technical requirements

  • Chrome browser
  • DataRobot Automated Machine Learning — If you don’t have access to the application, please sign up for our free trial: datarobot.com/trial.



Action Items45 min

  • Complete the Lab
  • Resources
  • Datasets