Classifier Combination for Capturing Musical Variation

At its heart, music information retrieval is characterized by the need to find the similarity between pieces of music. However, “similar” does not mean “the same”. Therefore, techniques for approximate matching are crucial to the development of good music information retrieval systems. Yet as one increases the level of approximation, one finds not only additional similar, relevant music, but also a larger number of not-as-similar, non-relevant music. The purpose of this work is to show that if two different retrieval systems do approximate matching in different manners, and both give decent results, they can be combined to give results better than either system individually. One need not sacrifice accuracy for the sake of flexibility.