TimeCLRS

This code was used in my master's degree experiment at Kyunggi University, and was written using RecBole as a reference.

Experiments Logs

Step 1: Justify time attributes in your recommendation system

Wandb

Step 2: Weighting the forward process to improve accuracy

Wandb

Step 3: Avoiding Precision/Recall Tradeoffs with Continual Learning

Wandb

Experimental Environments

Docker hub

Citation

If you utilize this repository, models or data in a downstream project, please consider citing it with:

@misc{TimeCLRS,
  author = {Sangmin Lee},
  title = {TimeCLRS: Improving user content prediction accuracy with continuous learning in time-based recommendation systems},
  year = {2023},
  publisher = {},
  journal = {},
  howpublished = {\url{https://github.com/d9249/TimeCLRS}},
}