Algorithm Can Guess Next Big Music Hit

by Laetitia Drevet 

If you wonder what this summer’s popular tune will be, why not ask your computer: a brand new algorithm might be able to guess it.

Analysing more than 500,000 songs released in the UK over the past three decades, researchers from the University of California, Irvine, have discovered a formula capable of predicting success in contemporary music.

The team matched the songs with their acoustic characteristics and musical features, such as happiness, sadness and danceability. They consider a song successful if it has made it to the UK top 100 charts.

Researcher Natalia Komarova said:

“It turned out that successful songs exhibit somewhat different trends and have quantifiably different features. They are on average significantly happier, more danceable, less relaxed and less aggressive, compared to other songs.”

Only four per cent of all the melodies are considered successful and they seem to follow their own trend. According to the study, an increasing amount of sad and emotional music is released every year.

And if women are increasingly present in the music industry, it is even more prominent when it comes to success: the algorithm points out that their tunes are nowadays more likely to become hits than men’s.  

In order to evaluate the algorithm’s reliability, the researchers tested it on more than a thousand songs released in 2014, making it guess whether or not these melodies had made it to the charts at the time.  The algorithm passed the test with honours: it was right 75 per cent of the time, raising to 85 per cent when the researchers added the “superstar variable”, that is if the singer had a song on the top 100 charts in the prior five years.

But despite these good results, the algorithm does not sound that appealing to some people in the industry.  

Tom Faber, a London-based DJ and promoter, said: “There are thousands different reasons that can drive someone to make music, and success is only one of them,“ he said.

For now, the algorithm does not pretend to identify more than trends. If it is able to list the ingredients of success, the final result cannot be guaranteed.  

“There is still more to music than mathematics can predict,” said Komarova.


Featured Image by geralt on Pixabay

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your 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

A Website.

Up ↑

%d bloggers like this: