How can I strengthen my machine learning model?


There are several methods to try if you want to refine and strengthen a machine learning model. Leverage the EDA Report and Results Summary Report to figure out what will work best for your model.

  • Recipe Reconfiguration
    • Increase or decrease history training intervals
    • Increase or decrease how far in the future you are predicting
    • Change the scope of what core attributes are being tried by your model to include more features or better features
    • Manually exclude bad columns from being included in your model
    • Create and include more or better generative attributes
  • Create separate models for different parts of your organization
  • Hone in on a subset of your population when building your model
  • Adjust your global settings to allow more columns to be tried
  • Perform per column interventions
    • Null Fill columns that are being dropped due to missing data
    • Adjust the droppability to allow the model to try columns that are being automatically dropped due to global settings
  • Configure the dimensionality reduction to increase or decrease the amount of columns your model includes
  • Work with your CS Lead and Data engineer to bring in new data such as compensation data, survey data, or performance data

Check out this help content about Refining a Machine Learning Model in One AI for more information and specific instructions. 


Was this article helpful?

0 out of 0 found this helpful



Please sign in to leave a comment.