|
Hit Song Science is a term coined by Mike McCready and trademarked by the company he co-founded, Polyphonic HMI. It concerns the possibility of predicting whether a song will be a hit, prior to its distribution using automated means such as machine learning software. ==Scientific background== The scientific nature of Hit Song Science is a subject of debate in the music information retrieval (MIR) community. Early studies claimed that using machine learning techniques could capture some information from audio signals and lyrics that would explain〔Dhanaraj, R. and Logan, B. Automatic Prediction of Hit Songs, Proc. of Ismir 2005, London, UK〕 popularity. However, a larger-scale evaluation〔Pachet, F. and Roy, P. (2008) (Hit Song Science is Not Yet a Science ). Proceedings of Ismir 2008, pages 355-360, Philadelphia, USA〕 contradicts the claims of “Hit Song Science”, i.e. that the popularity of a music title can be learned effectively from known audio features. Many other reasons, including the well known "cumulative advantage" or preferential attachment effects deeply contradicts〔Pachet, F. (2011) Hit Song Science. () In Tao, Tzanetakis & Ogihara, editor, Music Data Mining, CRC Press/Chapman Hall〕 the possibility of practical applications. Nevertheless, automatic prediction techniques are the basis of hit counseling businesses (HSS Technology). 抄文引用元・出典: フリー百科事典『 ウィキペディア（Wikipedia）』 ■ウィキペディアで「Hit Song Science」の詳細全文を読む スポンサード リンク
|