At the Ratemaking and Product Management (RPM) seminar of the Casualty Actuarial Society (CAS) in San Diego last month, Linda Brobeck, Peggy Brinkmann, and yours truly gave a concurrent session on a machine learning technique called decision tree analysis. See the DSPA-2 session — and other sessions — at this link. Technical issues precluded showing the video of installing and running Rattle within RStudio. Thus this post.
As Peggy notes, installing and running Rattle takes only three “commands” in R:
install.packages(“rattle”, dependencies=c(‘Depends’, “Suggests”))
Although not necessary, the “Suggests” option above can avoid Rattle’s annoying requests for more packages as you click through its GUI; the downside is the time to download and install about 900 packages. Below (“Read more >>”) is a link to a video of me running through the three lines above:
For those whose IT departments restrict installing R and RStudio on company equipment, this presentation used the software on a USB drive. The first minute or so of the video shows me setting up the shortcut on the USB drive so RStudio runs without a bunch of confusing questions, especially for a newbie. If you have R and RStudio installed on your computer already, you can fast forward through the first minute.
Peggy put her three lines of code into a script called ‘Rattle.R’. That’s the file I double-click around the 1:10 mark. You can see the three lines of the script in the upper left pane of RStudio.
I run the first line starting at 8:17 am, which finishes at 9:32 am (the 1:42 mark in the video). Plenty of time to read the newspaper.
Then I run the second line which, despite the “Depends” and “Suggests” settings above, has to download and install yet another package, GTK+. (There are reasons why that’s necessary, but not important right now.) That takes just a few minutes.
Then I run the third line, which fires up the Rattle user interface (GUI). It’s a little hard to see, but the Rattle GUI shows up as another process running in the taskbar.
At this point you can start using rattle/rpart per Peggy’s presentation. The links to our three presentations are reproduced here for convenience:
To leave a comment
for the author, please follow the link and comment on their blog: triKnowBits
offers daily e-mail updates
news and tutorials
on topics such as: Data science
, Big Data, R jobs
, visualization (ggplot2
), programming (RStudio
, Web Scraping
) statistics (regression
, time series
) and more...
If you got this far, why not subscribe for updates
from the site? Choose your flavor: e-mail
, or facebook