Wrow, Wcol

[snsatoshi]

How did you generate Wrow and Wcol that are contained in mat file ?

[fmonti]

As stated in the paper:

For the MovieLens dataset we constructed the user and item (movie) graphs as unweighted 10-nearest
neighbor graphs in the space of user and movie features, respectively. For Flixster, the user and item graphs were constructed from the scores of the original matrix (so still it's a nearest neighbors graph). For the Douban dataset, we used only the user graph (i.e. the provided social network of the user). For the YahooMusic dataset, we used only the item graph, constructed with unweighted 10-nearest neighbors in the space of item features (artists, albums, and genres).

[jiaxinwoo]

Hi,
For MovieLens dataset，what exactly are the user and movie features?

[fmonti]

You can find all the needed information in the dataset README:

http://files.grouplens.org/datasets/movielens/ml-100k-README.txt

[fmonti]

We used 10-NNs only when the original graph was not provided.