Direct and Inverse Inference in Music Databases: How to Make a Song Funk?

We propose an algorithm for exploiting statisticalproperties of large-scale metadata databases about musictitles to answer musicological queries. We introduce twoinference schemes called “direct” and “inverse” inference,based on an efficient implementation of a kernelregression approach. We describe an evaluationexperiment conducted on a large-scale database of fine-grained musical metadata. We use this database to trainthe direct inference […]