Applying TF-IDF to book descriptions
PyBooks has collected several book descriptions and wants to identify important words within them using the TF-IDF encoding technique. By doing this, they hope to gain more insights into the unique attributes of each book to help with their book recommendation system.
The following packages have been imported for you: torch
, torchtext
.
This exercise is part of the course
Deep Learning for Text with PyTorch
Exercise instructions
- Import the
TfidfVectorizer
class fromsklearn.feature_extraction.text
that converts a collection of raw documents to a matrix of TF-IDF features. - Instantiate an object of this class, then use this object to encode the
descriptions
into a TF-IDF matrix of vectors. - Retrieve and display the first five feature names from the
vectorizer
and encoded vectors fromtfidf_encoded_descriptions
.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Importing TF-IDF from sklearn
from sklearn.feature_extraction.text import ____
# Initialize TF-IDF encoding vectorizer
vectorizer = ____()
tfidf_encoded_descriptions = vectorizer.____(descriptions)
# Extract and print the first five features
print(____.get_feature_names_out()[:5])
print(____.toarray()[0, :5])