Installation: at the Julia REPL, Pkg.add("Surprise")
Reporting Issues and Contributing: See CONTRIBUTING.md
The main aim is to create a framework that facilitates the study of recommender systems in Julia.
julia> using Persa
julia> using DatasetsCF
julia> using Surprise
julia> dataset = DatasetsCF.MovieLens();
julia> model = Surprise.IRSVD(dataset);
julia> Persa.train!(model, dataset)
julia> model[1,1]
Rating: 4.010861934456679 (4)
List of collaborative algorithms:
| Algorithm | Description |
|---|---|
| KNNBasic | A basic KNN algorithm. |
| KNNBaseline | A basic KNN algorithm but using a baseline factor. |
| KNNWithMeans | A basic KNN algorithm but using user or item mean. |
| SlopeOne | SlopeOne algorithm. |
| RSVD | Regulared SVD. The algorithm is also known as SVD by Funk. |
| IRSVD | Improved Regulared SVD. Extension of RSVD algorithm adding the user and item bias. |