RSNMF recommender for rating prediction

[paraschakis]

Luo et al. (2014): "An efficient non-negative matrix-factorization-based
approach to collaborative filtering for recommender systems".
IEEE Transaction and Industrial Informatics, Vol. 10, No. 2, 2014.

[zenogantner]

Thank you for the pull requests.
Do you have measurement results (RMSE, MAE) on some public dataset like MovieLens?

[paraschakis]

Yep, finally managed 😊 You have already seen the results on MovieLens, and when I ran it on my e-commerce datasets it didn’t produce good results, perhaps because my datasets were binary. I haven’t tested it on other public datasets, so you might want to check the original paper. If it’s a fair performer, perhaps it’s worth to keep it just to enrich your collection of algorithms… but you know better.

Cheers,

Dimitris

From: Zeno Gantner
Sent: ‎Monday‎, ‎4‎ ‎January‎ ‎2016 ‎18‎:‎23
To: zenogantner/MyMediaLite
Cc: Dimitris Paraschakis

Thank you for the pull requests.
Do you have measurement results (RMSE, MAE) on some public dataset like MovieLens?
Is it really stronger than standard MF?

—
Reply to this email directly or view it on GitHub.