Project 3 Summary
This week I finished Project 3, which was a project that attempted to model the number of shares an internet article would get based on different predictor variables. I had one partner for this project and I think we worked well together.
My partner and I met briefly prior to beginning the project. I don’t know that there is anything I would have done differently with the project, although it took a little while to get going. The first part of the project was to subset the data based on the data channel that was assigned to the article. I found this part to be fairly easy by adding a new column to designate the column that the data came from. The second part that was trickier was to automate the analysis based on the input parameter (params$channel). I decided to create a function to look at the data channel (indicated by params$channel) and then assigned only that dataset to the active data set for the analysis. The rest of the project was fairly straightforward, although the final rendering was quite long with all of the models that had to run.
My big takeaways from this project is that being able to create reports automatically is a very powerful tool that I am happy to add to my toolbox. In this case our modeling was pretty absymal but I can see how powerful it could be. I thought it was really incredible how easy it was to create the reports once we had things set up properly.
You can see our final reports in the README.md file located here.
Our whole repository for this project is located here.