Tracking Emotions From Song Lyrics: Analyzing 30 Years of K-Pop Hits

Abstract

Emotions that are shared by a large number of people could broadly impact affective experiences at the individual level. Here, we used text mining on popular song lyrics—a cultural product that has been suggested to mirror emotions that many members of a society value and prefer—to track the changes in emotions over time. Morpheme frequency analysis and structural topic modeling on 2,962 hit K-pop songs from 1990 to 2019 showed converging evidence for increased positive emotional content and decreased negative emotional content embedded within the lyrics. This pattern of temporal shift in emotions aligned with rapid changes in South Korea in the past 30 years, notably a rise in individualism and ego orientation in a traditionally collectivistic culture, as well as economic growth. More generally, this study illustrates a strategy for tracking emotions that people value and prefer from large natural language data, supplementing existing methods such as self-reported surveys and laboratory experiments.