I created a pseudo-SVD++(3) model (mf_time.cpp in my framework) - one without the |N(u)|^(-.5)*Yi feedback term, and also without the alpha_u*dev_u_hat term (which I found to make the RMSE worse). However the one I initially wrote had a bug in the training and use of the alpha_u_k*dev_u_hat...

The Kadri Framework is meant for the analysis of the Netflix data set. It is based on Icefox's Netflix Recommender Framework. It uses Icefox's framework to build the core data files, and expands on it to allow pre-processing, the use of dates, and blending. Requirements: Qt LAPACK 64-bit...

Source code implementing the k-NN K Nearest Neighbors algorithm can be found at this post. It implements Icefox's framework, details of which are in this thread.