Fist, we have the number of lines, words and characters in the news file and an example.

I will explore this file to get the word frequency distribution and get the most commons, to do that I will take the 10% of the text to do an initial exploratory analysis. Keep in mind this file is taken of news.

## [1] "Lines: 1010242"

## [1] "Words: 34372598"

## [1] "Characters: 203791405"

## [1] "Example: Of course, Paul was 20 as a ro ..."

First, I took only those words that appear more than 5 times, the first chart shows us the histogram of frequency of the 95% of the words with lower occurrence. The second one is a word cloud of the top 1%, here we can see the most common word ‘said’ because this file is about news it’s not surprising.

## [1] "Most common word: said"

Repeating the same process applied to the news file, we get:

## [1] "Lines: 2360148"

## [1] "Words: 30373832"

## [1] "Characters: 162385042"

## [1] "Example: just wanted to thank you & ask ..."

We can see the most common words are: just, like, get, love and good. Also, in the word cloud we can see expressions widely used in social networks like haha and lol.

Repeating the same process applied to the news file, we get:

## [1] "Lines: 899288"

## [1] "Words: 37334441"

## [1] "Characters: 208361438"

## [1] "Example: The bruschetta however, missed ..."

In the blogs file, the most frequent words are: one, will, just, like and can, words very used to write about experiences and stories.