Week 8 practice

Text analytics

In Week 8 Workshop we are using text analytics and classification models to predict whether YouTube comments are HAM or SPAM.

Word Clouds

A very simple way of analysing text is to count occurances of words. We see word clouds a lot! How useful are they?

Look at the data cleaning steps required.

First, load your data:

Load in your ‘corpus’ or document

Is this more useful?

Could you improve it further?

Try commenting out the line that removes stopwords