Build the user profiles
You are now able to generate suggestions for similar items based on their labeled features or based on their descriptions. But sometimes finding similar items might not be enough. In the next exercises, you will work through how one could create recommendations based on a user and all the items they liked as opposed to a singular item. You will first generate a profile for a user by aggregating all of the movies they have previously enjoyed.
The tfidf_summary_df you have been working on in the last few exercises has been loaded for you. This contains a row per movie with their titles as the index and a column for each feature containing their respective TF-IDF score.
This exercise is part of the course
Building Recommendation Engines in Python
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
list_of_movies_enjoyed = ['Captain America: The First Avenger', 'Green Lantern', 'The Avengers']
# Create a subset of only the movies the user has enjoyed
movies_enjoyed_df = tfidf_summary_df.____(____)
# Inspect the DataFrame
print(movies_enjoyed_df)