...

/

MD Implementation Steps: 9 and 10

MD Implementation Steps: 9 and 10

This lesson will finish going over the implementation steps (9-10) of the model design.

9) Evaluate

The next step of the model design process is to evaluate the results. The evaluation method will depend on the scope of your model. Specifically, this depends on whether it is a classification or regression model. In classification, the common evaluation methods are the confusion matrix, classification report, and accuracy score.

Accuracy=NumberPredictedCorrectlyNumberofCasesAccuracy = \frac{Number Predicted Correctly}{Number of Cases}

Accuracy score

This is a simple metric measuring how many cases the model classified correctly divided by the full number of cases. If all predictions are correct, the accuracy score is 1.0, and 0 if all cases are mispredicted.

While accuracy alone is normally a reliable performance metric, it may hide a lopsided number of false positives or negatives. This isn’t a problem if there are a balanced number of false positives and false negatives, but also isn’t something you can ascertain using accuracy alone. Thus, we turn to the following two evaluation methods.

Follow the provided link if you need more detailed explanations of Accuracy Score.

Confusion matrix

A confusion matrix (also known as an error matrix) is a simple table that summarizes the ...

Access this course and 1400+ top-rated courses and projects.