Music Charts for Approximating Everyday Emotions: A Dataset of Daily Charts with Music Features from 106 Cities

The music we listen to often reflects our emotional state. Accordingly, a city's music charts may reflect consistencies and patterns in the collective emotional experiences of its residents. In this paper, we report a dataset of arousal and valence-related music features collected from songs on Top-25 charts from 106 cities around the world, over a period of 158 days. Chart information was obtained from the Apple Music platform, and music features were obtained from Spotify. We visualize daily fluctuations of features corresponding to intensity arousal (energy, loudness), rhythmic arousal (danceability, tempo) and valence, highlighting trends in nine major cities across six continents, and conduct k-means clustering of all 106 cities from these five music features. Preliminary results suggest consistencies between our music feature data and previous literature, in reflecting cultural differences for preferred arousal levels in music. We also discuss possible implications of our paper and data in the use of music streaming charts as indicators for collective emotional experience.