The program look at 23 separate variables such as loudness, danceability and complexity of harmonies, it's had a 60% success rate when using the Top 40 charts from the past 50 years as data.
"The goal was to find out if we could come up with an equation that distinguishes between a hit and something that dangles at the bottom of the charts," said Dr Tijl De Bie, a senior lecturer in artificial intelligence at Bristol, who heads the research team.
Dr Bie said the equation was developed using the publicly available data about songs in the UK top 40 since 1961. For each week in that long history, the equation was tested with new releases to see if it could predict where that song would get in the chart.
"At every moment in time the equation can be different because we only took into account past data," Dr De Bie told the BBC.
Machine learning techniques were used to help the equation learn about the relative importance of all the elements that make up a pop song. The result, he said, was an equation that is right more often than it is wrong.
You can read the entire article here.