Exercise

Explore gender data

The data gender contains the Weight, the Height and the BMI indices of 10,000 people. The original data has a Gender label for 5,000 people that identify themselves as females and the other 5,000 as males. The labels will be useful later for testing how well the clustering performs against the real labels. However, in this subset of the dataset, the labels are not provided.

The data gender_with_probs also contains the probabilities of each data point belonging to a cluster. Since we are interested in two clusters, probabilities near 1 refer to one cluster and near 0 to the other.

The aim of this exercise is to have a glance at how a typical clustering dataset looks like before and after clustering.

Instructions

100 XP
  • Use the function head to look at the first 6 observations of gender.
  • Use the function head to look at the first 6 observations of gender_with_probs.
  • Make a scatterplot with Weight in the x-axis and BMI in the y-axis. Colour the dots by their probability.