As we can see, Bob Dylan was the most prolific in the 1960s and his song output decreased over the decades. However, in the 2000s, he made a comeback.
Unsurprising for a popular song writer, in general, but perhaps surprising for Bob Dylan, in particular, “love” is the most frequently used (non-stop) word in his lyrics.
These results are perhaps the most interesting. Although “love” is the most commonly used word overall, “time” is the most frequently used word in Dylan’s lyrics in the 1960s, 1970s, and 1980s and is in the top 8 during the 1990s and 2000s. (Although “time” is a very commonly-used word in general) In the 2010s, neither “love” nor “time” is in the top 8 and things get a bit dark.
Now we look at the most popular words for each album.
Lexical diversity has increased slightly over the years in Bob Dylan’s lyrics. In other words, he’s been using a wider range of words.
Lexical density has decreased slightly over the years. That is, his songs “say less” over time. Given the high density and complexity of his lyrics from 1964–1966, this is probably not surprising. However, lexical density has been making a comeback in his most recent songs.
See Wikipedia for a definition of TF-IDF. Using TF-IDF on albums is quite informative. We see which words are important on a particular album.
This article is highly indebted to the techniques used in “Lyric Analysis with NLP & Machine Learning with R” written by Debbie Liske.
My code is available in the bob-dylan-haikus project.