The science behind dance hit prediction
What makes a dance song a hit? Our team at the ANT/OR research group and the Applied Data Mining Group from the University of Antwerp has delved into this question. Dorien Herremans, David Martens and Kenneth Sörensen applied data mining techniques to develop an accurate classification model that can distinguish top 10 verus top 30-40 hit listings. The online app allows you to upload a dance song and calculate the probability that this song will be a dance hit.
This research was presented at the Workshop for Music and Machine Learning in Prague, 23th september 2013. An abstract of the presentation:
With annual investments of several billions of dollars worldwide, record companies can benefit tremendously by gaining insight into what actually makes a hit song. This question is tackled in this research by focussing on the dance hit song problem prediction problem. A database of dance hit songs from 1985 until 2013 is built, including basic musical features, as well as more advanced features that capture a temporal aspect. Different classifiers are used to build and test dance hit prediction models. The resulting model has a good performance when predicting whether a song is a “top 10” dance hit versus a lower listed position.
A full paper describing all technical details can be found here:
Recent lookups
Title | Artist | Hit potential |
---|---|---|
Hey Brother | Avicii | 0.79075215208375 |
0.33349905702209 | ||
0.32063198511117 | ||
0.66343603829679 | ||
0.8540096861118 | ||
0.54892353607647 | ||
0.75902066682094 | ||
0.3373527009869 | ||
0.52561368189125 | ||
Hey Brother | Avicii | 0.79075215208375 |
0.94791973764971 | ||
0.44413911893519 | ||
0.61324589313231 | ||
0.8042472949357 | ||
0.9230797419871 | ||
ADRII | 0.37835287437733 | |
Miss You | Armin Van Buuren | 0.78938822048411 |
0.57979223543412 | ||
0.75902066682094 | ||
0.86423662938825 | ||
0.7137314075794 |