Text Analysis of Popular Music from the Past Decade

Text Analysis of Popular Music from the Past Decade



Jesse P. Devitt


Presentation: 3:00-3:20, Kennedy Union 211



Millions of songs, ranging from a multitude of genres, are released every year. And now, due to popular streaming applications, music is more accessible than ever before. This project explores the impact of lyrics on a song's appeal. Utilizing text and sentiment analysis, it examines the relationship between lyrics and song popularity. The data for this analysis comes from Genius. This popular online, user-centric media company provides the lyrics to millions of popular songs and the space for users to annotate and analyze said lyrics. Focusing on songs from the last decade (2010-), we aim to highlight the most frequent words and word pairs within specific genres through word clouds and bigram analysis, respectively. In addition, some of the more popular artists from the decade were isolated and analyzed via sentiment analysis of their lyrics, hoping to find correlations to song popularity via site views. This project possesses the goal of understanding the complexities within the modern-day music scene. With all the advancements in artificial intelligence, it's within the realm of possibility that a once artist-centric industry may shift over to big data, as so many other industries have in recent years.

Publication Date


Project Designation

Capstone Project

Primary Advisor

Ying-Ju Chen

Primary Advisor's Department



Stander Symposium, College of Arts and Sciences

Institutional Learning Goals

Scholarship; Critical Evaluation of Our Times; Diversity

Text Analysis of Popular Music from the Past Decade