15.071 | Spring 2017 | Graduate

The Analytics Edge

6.2 Recommendations Worth a Million: An Introduction to Clustering

6.2 Recommendations Worth a Million: An Introduction to Clustering

Quick Question

Run the cutree function again to create the cluster groups, but this time pick k = 2 clusters. It turns out that the algorithm groups all of the movies that only belong to one specific genre in one cluster (cluster 2), and puts all of the other movies in the other cluster (cluster 1). What is the genre that all of the movies in cluster 2 belong to?

Explanation You can redo the cluster grouping with just two clusters by running the following command: clusterGroups = cutree(clusterMovies, k = 2) Then, by using the tapply function just like we did in the video, you can see the average value in each genre and cluster. It turns out that all of the movies in the second cluster belong to the drama genre. Alternatively, you can use colMeans or lapply as explained below Video 7.

Course Info

As Taught In
Spring 2017
Level
Learning Resource Types
Lecture Videos
Lecture Notes
Problem Sets with Solutions