If you would like, you can view the written report our group submitted, and the slides that were presented to the class.
Here is the R code we used to do preliminary cleanup of the data set.
Chase made a number of visualizations for the project, which are also seen in the paper and presentation slides.
This is the program we used to grab additional data from Kickstarter using the existing project names and ID
numbers from our Kaggle data. To be able to make an appropriate request, the scraper itself is implemented in
Python and is called from R using the
The Python program uses a Virtualenv, so if you are on a UNIX system, use the .tar.gz to make sure the permissions are maintained correctly.
This set of programs generated a word cloud showing the frequency of word use in the "blurb" fields of the Kickstarter projects.
A small utility program for merging two versions of the dataframe together. While the scraping program (above) was running, other members of the group had made advancements in the dataframe, so the two branches of changes needed to be merged. This program accomplishes that.
This is the source for the model prediction app in the demo above. The archive contains the data files, all the source needed to generate the model, as well as the source code for turining it into the Shiny app itself.