This was the collaborative filtering system I developed for my masters degree. It implements a series of methods for selecting items to present to users to rate in an attempt to alleviate the cold-start problem.
The prediction model is that published by Koren and Bell [1], who formed part of the team that won the Netflix Prize.
Building the project is as simple as issuing a make command. However, due to the binary data required to run the program, which I'm unable to distribute, the program will not run.
Warning: This project was a single developer project, hence some of the commit messages are lacking detail.
[1]: Scalable Collaborative Filtering with Jointly Derived Neighborhood Interpolation Weights. ICDM 2007.