Music Popularity: Metrics, Characteristics, and Audio-Based Prediction
This addresses the need for better understanding and prediction of music popularity for artists and the music industry, but it is incremental as it builds on existing acoustic feature methods.
The paper tackled the problem of defining, characterizing, and predicting music popularity by proposing eight metrics and analyzing real-world chart data, and found that predicting popularity from audio signals is feasible with significant improvement over random chance using complexity and MFCC features.
Understanding music popularity is important not only for the artists who create and perform music but also for the music-related industry. It has not been studied well how music popularity can be defined, what its characteristics are, and whether it can be predicted, which are addressed in this paper. We first define eight popularity metrics to cover multiple aspects of popularity. Then, the analysis of each popularity metric is conducted with long-term real-world chart data to deeply understand the characteristics of music popularity in the real world. We also build classification models for predicting popularity metrics using acoustic data. In particular, we focus on evaluating features describing music complexity together with other conventional acoustic features including MPEG-7 and Mel-frequency cepstral coefficient (MFCC) features. The results show that, although room still exists for improvement, it is feasible to predict the popularity metrics of a song significantly better than random chance based on its audio signal, particularly using both the complexity and MFCC features.