Binomial distribution beats multinomial distribution test data
298
|
---|
We experimented with EachMovie dataset2 containing ratings for films. User ratings are expressed on a 6-point scale. We selected a set of 100 most rated films. The number of users that rated at least one film from the selected set was 60, 895.
We partitioned the set of ratings into two sets – training and test
sets. The training set | is used to train the models and visualize the |
---|
sets to compare the models.
structure on ordered rating values | Models with unconstrained latent |
---|
|
|
---|