Music is a collection of notes. When those notes are played and recorded, the medium used is most often digital in nature. At it’s essence, digital data is just a combination of ones and zeroes stored in electronic fashion. Thus the digital information that is generated by musical recordings can be evaluated in a similar manner to all other types of digital data. For example, big data analysis frequently involves lots of pattern matching. Music data is an ideal candidate for discovering patterns. Identifying a pattern in the melody, harmony, rhythm, bass, etc. that a particular user finds appealing enables song recommendation software to recommend similar music that the user may not be aware of.
Identifying patterns can provide other benefits. A big problem that the music industry faces is piracy of recorded material. Analyze has written software that helps identify copyright violations on web-posted recordings. For example, YouTube, Dailymotion, Vimeo, and Vevo have many recordings posted online that can be accessed for free by anyone. Are the artists and publishers being fairly compensated for their work? Do they even know how much of their work is being made available for free? Analyze has applied big data pattern-matching techniques against open-source music repositories to identify online musical recordings. When the user indicates the musical recording he wants evaluated, the resulting report provides concrete evidence of musical usage of that recording. Pattern matching at its finest. See this link to view the product called AnalyzeStreamTrack. So, big data can contribute to the music industry just as it does too so many other industries.
A good article for a further read on how another company called Pandora analyzes music to offer customized recommendations is found at this link.