The center for jazz studies at Columbia university has launched a blog as part of the music information retrieval initiative (MIR) initiative as seen in this
article.
MIR(Music information retrieval) is an interdisciplinary science of retrieving information from music. MIR is used to categorize music for retrieval. Music Information Retrieval (MIR) is a young field that applies tools from machine
learning and signal processing to obtain information about musical items. Despite significant work with pop and classical music, jazz remains almost completely unstressed by the MIR community. Features like rhythmic/harmonic complexity and improvisational structure mean that jazz poses a set of novel problems to explore in the context of these tools.
The vision of Music information retrieval is to build a recommend er artificial intelligence system which does not rely on human experts for music classification but has a retrieval algorithm in place which will analyze data in the music record to facilitate music retrieval.
An important step in analyzing data is feature extraction. The music data is analyzed with respect to features such as beats per minute,main pitch,harmony,chords to be fed into machine learning algorithms to perform automatic classification of music data in terms of mood of the piece in addition to factual data such as such as identifying
the artist,place of origin. This method of information organization allows us to learn intrinsic characteristics to support retrieval methods such as semantic retrieval based on pattern matching and even retrieval by humming, an advanced form of pattern matching. The retrieval can also be made based on derived characteristics of the music record such as whether it might be recommended for a particular occasion based on the mood of the song. The process would facilitate a lesser trod path of information retrieval for music data through web search for recommendation engines or even query by humming in smartphones. The uses of it are myriad and its certainly interesting to look at information retrieval for an entity as different as music !