Quality of a Classifier Model

MUJEEBUR RAHMAN 16 Reputation points
2020-09-15T12:25:51.413+00:00

I have been trying out a classifier model in ML studio, and I have got some results in the Visualize option. I can see the values for Precision, Recall, Accuracy, threshold and also the Confusion Matrix. Now my question is, which of the above mentioned parameters describes the quality of the model?

Azure Machine Learning
Azure Machine Learning
An Azure machine learning service for building and deploying models.
2,848 questions
0 comments

1 answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. Ramr-msft 17,736 Reputation points
    2020-09-16T13:36:07.547+00:00

    @MUJEEBUR RAHMAN Thanks, If your ML model has good performance metrics (you should decide which works best based on confusion matrix and which error can be handled and which should be reduced) then it is a great solution. All the metrics computed and displayed are on validation set and not on the original training set.

    1. Split the provided dataset in train / validation / test subsets (if chooses the additional test option)
    2. Use train and validation sets to find the best model
    3. Retrain best model with best algo and parameters using combined train + validation sets
    4. Evaluate on test set and check the results.

    If possible please share the classifier model that you are trying.

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.