This is our implementation for the paper:

Pan Li, and Alexander Tuzhilin. "Latent multi-criteria ratings for recommendations." Proceedings of the 13th ACM Conference on Recommender Systems. 2019. [Paper]

We provide the sample dataset as a showcase. You are always welcome to use our codes for your own dataset.

Please cite our RecSys'19 paper if you use our codes. Thanks!

Author: Pan Li (https://lpworld.github.io/)

We use PyTorch and Tensorflow as the backend.

• PyTorch version: '1.2.0'
• Tensorflow version: '1.4.0'

The instruction of commands has been clearly stated in the codes (see the parse_args function).

Run LatentMC:

python train_latentmc.py

or alternatively

python train_latentmc.py --cuda

This implementation uses some codes from Adversarially Regularized Autoencoders, Compress Word Embeddings and Surprise.

Last Update: 2020/02/17