SweetRS: Dataset for a recommender systems of sweets
This provides a new dataset for benchmarking recommender systems, which is incremental as it adds to existing resources without introducing novel methods.
The authors tackled the problem of benchmarking recommender systems by creating a dataset of 77 sweets with over 44,000 grades from 2,000 users, achieving 28% matrix coverage, and they benchmarked the Soft-Impute algorithm on this dataset.
Benchmarking recommender system and matrix completion algorithms could be greatly simplified if the entire matrix was known. We built a \url{sweetrs.org} platform with $77$ candies and sweets to rank. Over $2000$ users submitted over $44000$ grades resulting in a matrix with $28\%$ coverage. In this report, we give the full description of the environment and we benchmark the \textsc{Soft-Impute} algorithm on the dataset.