Hands-On Python Deep Learning for the Web

Bag of words

It is not possible for computers to directly process and work with text. Hence, all text must be converted into numbers before being fed into a machine learning model. The process of changing text to an array of numbers, such that it is possible to retrieve the most important pieces of the original text from the converted text at any point in time, is known as feature extraction or encoding. Bag of Words (BoW) is one popular and simple technique used to perform feature extraction on text.

The steps associated with a BoW implementation are as follows:

1. Extract all the unique words from the document.
2. Create a single vector with all the unique words in the document.
3. Convert each document into a Boolean array based on whether any ...