Science can possibly predict if a song will scale the hit charts by using a machine learning algorithms, research shows.

A study led by Dr Tijl de Bie of the University of Bristol's Intelligent Systems Laboratory in the Faculty of Engineering found that predicting how a song will fare may be feasible using the machine.

The team looked at the official UK top 40 singles chart over the past 50 years to distinguish the most popular songs from less popular singles. They then devised a "hit potential equation" that scores a song according to its audio features.

The researchers used musical features such as tempo, time signature, song duration and volume. They also computed more detailed analyses of the songs such as harmonic simplicity, how simple the chord sequence is, and non-harmonicity, how 'noisy' the song is.

From a list of UK hits for a certain time and measuring their audio features, the researchers measured how important each of the 23 features was and computed a score for a song. The equation works by looking at all the scores generated by each of the song.

The team found they could classify a song into a 'hit' or 'not hit' based on its score, with an accuracy of 60 per cent as to whether a song will make it to the top five, or if it will rise above position 30 on the UK top 40 singles chart.

"Musical tastes evolve, which means our 'hit potential equation' needs to evolve as well. Indeed, we have found the hit potential of a song depends on the era. This may be due to the varying dominant music style, culture and environment," said De Bie.

The research findings were presented at an international conference, the MML 2011 4th International Workshop on Machine Learning and Music: Learning from Musical Structure held in conjunction with the 25th Annual Conference on Neural Information Processing Systems held in Sierra Nevada, Spain Saturday.