Finding similar users using collaborative filtering
Collaborative filtering refers to the process of identifying patterns among the objects in a dataset in order to decide about a new object. In the context of recommendation engines, collaborative filtering is used to provide recommendations by looking at similar users in the dataset.
By collecting the preferences of different users in the dataset, we collaborate that information to filter the users. Hence the name collaborative filtering.
The assumption here is that if two people have similar ratings for a set of movies, then their choices for a set of new unknown movies would be similar too. By identifying patterns in those common movies, predictions can be made about new movies. In the previous section, we learned how to compare different users in the dataset. The scoring techniques discussed will now be used to find similar users in the dataset. Collaborative filtering algorithms can be parallelized and...