MovieLens is a movie recommendation website. It provides a popular dataset that many people have built tutorials on.
As a university assignement (ProjetSid.pdf), Giovanni Zanitti and I had to figure out if clusters of individuals could be extracted. The whole report is named Rapport_Zanitti_Halford.pdf and details the procedures we took and the choices we made.
The data munging was performed with Python (and the pandas module) whilst the statistical analysis was done with R (thanks to the FactoMineR package).
Here are some plot examples:
We also produced a map of the United States showing the average scores given to the movies by the users, but it was not included in the report:
For more information our contacts are:
maxhalford25@gmail.comgiovanni.zanitti@gmail.com