0

I am working on wine quality dataset for predicting if the quality of wine is good or bad.

I have used multiple classification models and calculated their accuracy/Precision/Recall score as shown below

Accuracy\Precision\Recall

However , I can not rely on accuracy score as data is imbalance i.e. Data with class Bad(1382 rows) is greater then class Good

Which model should I prefer based on above precision and recall score and why?

desertnaut
  • 57,590
  • 26
  • 140
  • 166
pankaj mishra
  • 2,555
  • 2
  • 17
  • 31
  • 1
    I’m voting to close this question because it is not about programming as defined in the [help] but about ML theory and/or methodology - please see the intro & **NOTE** in the `machine-learning` [tag info](https://stackoverflow.com/tags/machine-learning/info). – desertnaut Feb 25 '21 at 20:02
  • 1
    That said, the appropriate metric is normally dictated by the *business* problem we are trying to solve. – desertnaut Feb 25 '21 at 20:03
  • You may find [F1 score](https://en.wikipedia.org/wiki/F-score) useful, it is harmonic mean of precision and recall. – null Feb 25 '21 at 20:43

0 Answers0