The Art of R Programming

[1] 9162933

$of

[1] 11 21 42 56

$output

[1] 12 19 39 57

Yep, ordered from least to most frequent.

We can also do a plot of the most frequent words. I ran the following

code on an article on R in theNew York Times, “Data Analysts Captivated by

R’s Power,” from January 6, 2009.

> nyt <- findwords("nyt.txt")

Read 1011 items

> snyt <- freqwl(nyt)

> nwords <- length(ssnyt)

> barplot(ssnyt[round(0.9*nwords):nwords])

My goal was to plot the frequencies of the top 10 percent of the words in

the article. The results are shown in Figure 4-1.

Figure 4-1: Top word frequencies in an article about R

98 Chapter 4