English subtitles

← Friend Count - Data Analysis with R

Get Embed Code
5 Languages

Showing Revision 3 created 05/25/2016 by Udacity Robot.

  1. To create the histogram, we'll pass friend count to x
  2. inside of hue plot, and we'll pass pf into data.
  3. When I run the code, I can get my histogram. Looking
  4. at the plot, we can see that the data is squished on
  5. the left side of the graph, just like Maura's plot, and
  6. this graph extends all the way to the 5000 mark. This is
  7. what we call long tail data. And this data can be
  8. common for some user-level data. Most users have friend counts under 500,
  9. so we get really tall bins on the left. But there are
  10. a few users in our dataset with really high values. The higher
  11. values are closer to 5,000, which is the maximum
  12. number of friends a user can have. Moore was interested
  13. in examining the bulk of her data. And, it's the
  14. same for us. While, we may want to investigate some
  15. observations in this tail, we really want to examine our
  16. users with friend counts well below 1,000. All of these
  17. people here. To do that, we'll need to learn something
  18. else, in order to adjust our code, and our plot.