Performance Measures Of Predictive Models
By now, you should have a general sense of the types of predictive models, for example:
- Linear Regression
- Decision Trees
- ANN
If you’d not, feel free to check out a summary of commonly used predictive models below:
In this post, I would mostly be covering the performance measures of predictive models and the types to be used for different target types.
Depending on the outcome of our target, numeric or binary, we can apply the following performance measures for Classifiers (Binary targets) or Estimators (Numeric targets) to aid us in model evaluation and predicted results.
Some examples of Classifiers / Binary targets could include: 0 or 1, Yes or No, Purchase or No Purchase
Some examples of Estimators / Numeric targets could include Values which are not Binary and mutually exclusive.
Common performance measures for Classifiers
- Confusion Matrix
