Making recommendations with TF-IDF
In the last exercise you pre-calculated the similarity ratings between all movies in the dataset based on their plots transformed by TF-IDF. Now you will put these similarity ratings in a DataFrame for ease of use. Then you will use this new DataFrame to suggest a movie recommendation.
The cosine_similarity_array containing a matrix of the similarity values between all movies that you created in the last exercise has been loaded for you. The tfidf_summary_df DataFrame containing the movies and their TF-IDF features is also available.
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.
# Wrap the preloaded array in a DataFrame
cosine_similarity_df = pd.____(____, index=tfidf_summary_df.index, columns=tfidf_summary_df.index)
# Find the values for the movie Rio
cosine_similarity_series = ____.____['Rio']
# Sort these values highest to lowest
ordered_similarities = cosine_similarity_series.____(____)
# Print the results
print(____)