This is the analysis step. Some people would refer it as the actual data mining, or, in this case, text mining. For other people, we have been doing text mining for a long time now. Terminology aside, we have a clean and transformed dataset (clean_dt) that very much speaks for itself.

The features displayed by the tibble might be useful for some people already. It is for me, as I can drive my studies and seek some more R adventures. Yet, the analysis could be deepened with no troubles; data mining is never about getting enough knowledge, but about maximizing the amount of insight we can get given our computing and time constraints.

As you get skilled and experienced, you can get and deliver more out from it. In this section, we will depart from our clean dataset to do the following:

• Draw some descriptive analysis
• Learn cool ways to visualize this kind of data 
• Fit hierarchical clustering (and visualize it again!)

Hands on deck, sailor.