acm-header
Sign In

Communications of the ACM

ACM TechNews

AI Can Predict If Your Musical Taste Is More Ice T...or Vanilla Ice


View as: Print Mobile App Share:
Measuring a person's "musical sophistication."

A new system considers a persons listening behaviors, then uses machine learning algorithms and psychological models to infer their musical sophistication.

Credit: LuckyImages/Shutterstock

Researchers at Jonkoping University in Sweden and Maastricht University in the Netherlands have developed a system that considers a user's listening behaviors and uses machine learning algorithms and psychological models to infer their "musical sophistication."

Previous studies have shown that people with a higher degree of musical sophistication are more musically skilled, and engage in more musical behaviors, such as practicing an instrument or listening to a variety of musical genres.

The researchers collected data from Spotify's application programming interface, enabling access to users' playlists and audio features such as liveliness, time signatures, and artist popularity.

The data was fed into a neural network that predicted 61% of subjects' emotions and active musical engagement with a high degree of accuracy.

Compared to the baseline, the system was 95% accurate at predicting their emotions and 93% accurate at predicting active musical engagement.

From Venture Beat
View Full Article

 

Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account