This implies that the structure of lyrics is shifting towards containing more choruses than in the past, in turn contributing to higher repetitiveness of lyrics
The two main data sources for our investigation are last.fm and Genius. Given the nature and history of these platforms, in particular last.fm, the studied LFM-2b dataset is affected by community bias and popularity bias.
So, what we’re really looking at is that popular songs (those that withstood the test of time, not necessarily popularity on release) of yesteryear were more complicated. So, this basically just confirms that we only care about the good music from the past and we forget the boring and repetitive stuff. And not even necessarily more complicated, just fewer repetitions of the chorus. I’d care more about a comparison that controls for chorus repetition and chart position/popularity