Interesting concept, but I prefer movielens myself. Having good grades by people who like similar movies is imo a better predictor if I will be enjoying the mov

Interesting concept, but I prefer movielens myself. Having good grades by people who like similar movies is imo a better predictor if I will be enjoying the movie or not. How I'm feeling is of course important, but if I can see the short description of the movie I can decide for myself... Still, movielens dataset is available, maybe combination of the two approaches would be even better?