Let’s now see how LSA is used to generate vectors for the words in python.
In the previous lab, the number of documents was kept constant and the column size was reduced, where columns represented the terms.
However, in the following exercise, the number of terms is kept constant while the number of row(i.e. document) is reduced.
If you compare this code-lab with the one in LSA,, you will find that most of the code is similar except how tf-idf is used. This results in generating vectors for words in this lab, and vectors for documents in the previous lab.
In the next lecture, you will learn to create word vectors using word2vec in Python.