Skip to content

Releases: iemre/MRSR

Fix RMSE calculation

04 May 12:30
Compare
Choose a tag to compare

RMSE calculation was not applying the sqrt function to the result. Now fixed and tested.

MRSR v1.4

02 Jan 21:18
Compare
Choose a tag to compare
  • Minor fixes and improvements

MRSR v1.3

10 Dec 17:49
Compare
Choose a tag to compare

MRSR - Matlab Recommender System Research

22 Oct 11:39
Compare
Choose a tag to compare

MRSR is a set of MATLAB classes for recommender systems research.
The idea is to put all recommendation algorithms in one place, provide a flexible design for researchers to plug in their own recommendation algorithms, and make reliable comparisons by using common evaluation methods.

Matlab Recommender System Research

25 Jul 11:10
Compare
Choose a tag to compare
  • Stable release for MRSR.
  • The user can download the source code and start experimenting with item-based KNN and sparse coding based recommender systems. The user can add their own recommender system by extending the AbstractExperiment class. Evaluation is handled by the software for the user.
  • Evaluation metrics: MAE & RMSE, CPP, Recall, Precision, Personalisation. The user can add other metrics in AbstractExperiment.m.