Session Ready
Exercise

Checking for class imbalance

Classification problems where one decision class is overrepresented and the other ones are underrepresented can have a detrimental effect on your model fitting. Recognizing when you are in the presence of an imbalanced classification problem is the first step to adequately deal with this issue.

In this exercise, you are going to check whether the pulsar dataset suffers from class imbalance or not. This dataset labels a star as being of a pulsar type or not depending on eight numerical measurements from the star's integrated profile and its DM/SNR curve.

Then, you will proceed to partition the pulsar data into training and test sets. The dplyr and caret packages have been loaded for you.

Instructions 1/4
undefined XP
  • 1
  • 2
  • 3
  • 4
  • Glimpse at the pulsar data.