Domain Optimization for Classification

  • Updated

Classification using simClassify+ has extra parameters and features for optimizing results based upon a domain attribute. The domain attribute can be any REAL column and could represent money, effort, size, and so on. Typically, this would be used in fraud detection with the domain attribute being the amount of money in a transaction. In this example, this directs the metric learner to optimize not just for fraud transactions caught, but fraud money caught.

To use Domain Optimization in the user interface, first select the “Use Domain Property” slider. Two additional fields will appear on the find or enter parameters page, “Domain Column” and “Domain Importance Function”. In the “Domain Column” drop down, select the REAL or IGNORE column to optimize the results on. If choosing a column with an IGNORE spec type, that column must still be numeric to be used as the “Domain Column”. The “Domain Importance Function” field will determine how the optimization is applied in the learning process. In the drop down, there will be three options available to the user:

  • Global - optimization emphasis placed on both classes
  • Conservative - optimization emphasis placed on positive class only, conservative weighting approach
  • Aggressive - optimization emphasis placed on positive class only, aggressive weighting approach


Was this article helpful?

0 out of 0 found this helpful



Please sign in to leave a comment.