Accuracy

This is the second blog of series -“Evaluation Metric”. Hope you have idea about Confusion Matrix and terms involved with it-: TP, TN etc. If you want to look at it just go through this tutorial: http://blog.robofied.com/model-evaluation-metrics/

Let’s consider this confusion matrix :


By its name one can easily guess that Accuracy is used to measure how accurate our model i.e, how many positives are predicted positive and how many negatives are predicted negative.

Accuracy = (TP+TN)/(TP+TN+FP+FN)

From the above formula, we deduce-:

In numerator there are correct predictions made by model and in denominator, there is involvement of all the terms whether it is wrong or right.

Advantages of using Accuracy

  1. It is easy to use.
  2. It gives general overview of Accuracy.

Disadvantages of using Accuracy

  1. It gives bad results if data set is unbalanced i.e, equal no. of observations belong to each class.
  2. It is not judgemental  i.e, we cannot say higher accuracy means good model and vice-versa -: Sometimes it may be desirable to select a model with a lower accuracy because it has a greater predictive power on the problem.

Hope you enjoyed reading this blog.

This is for accuracy evaluation metric.

 

 

 

Leave a Comment

Your email address will not be published. Required fields are marked *