Emotion Mining and Topic Extraction on Song Lyrics

Emotion Mining and Topic Extraction on the song lyrics of 100 Top Billboard and 20 metal bands show clear similarity in sentiments and topics between artists that belong to the same genre. In two recent studies from the Department of Data Science and Knowledge Engineering of the Maastricht University, students aim to measure the similarity of artists with respect to emotions and topics of their songs based on their lyrics.

 

Sentiment mining is a technique that detects and analyzes human emotions towards events, people or other interests. The rise of social media offers many interesting applications for sentiment and emotion mining. Analyzing posts, comments, blogs, reviews, ratings, recommendations and other forms of online expression, offers valuable information for enterprises to position their products, identify new opportunities and manage their reputations. In (criminal) investigations, sentiments expressed in (electronically stored) communications like emails and chats, form a good starting point for further research.

Topic Extraction: Metal Bands Wonder about Time

In their project “Emotion Mining and Topic Extraction on Lyrics” Christoph Emunds and Richard Polzin analyzed 11.000 songs from 120 artists. Their research findings show that towards the more negatively connoted emotions anger, fear, sadness, and disgust, there are a lot of Hip Hop, Rap, and Metal artists. Music associated with the Pop and Mainstream genre is more connected to the positive emotions joy, surprise, trust, and anticipation.

1

Looking at the topics of the songs, the researchers find that rappers and hip-hop artists often sing about money, being rich and sexual intercourse. The second cluster is all about partying and having a good time, and feeling alive or real.

Metal bands form an exception. Their songs are in many cases associated with the topic time. And not like the pop artist about partying and having a great time, but rather about the meaning of time itself. The third cluster comprises everything about human relationships, from love to breakup.

In the earlier study “Information Retrieval and Text Mining in Lyrics” Torsten Schuster and Alexey Sibirts mined 228645 lyrics from 16518 artists.

2

Their objective was to clustered artists with similar lyrics and determine the emotional direction of these artists and their song. Their visualization also show rappers and hip hop-artist to be clustered around negative emotions like anger.

And Justin Bieber? He is expressing all emotions in equal amounts, meaning he is either a multi-talented artist or he is seriously confused. The choice is yours!

3
Emotion Mining on the Lyrics of Justin Bieber

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s