String-Based Methods for Tonal Harmony
David R. W. Sears
Abstract
Abstract Corpus studies employing string-based methods in music research often suffer from the contiguity fallacy - the assumption that note or chord events on the musical surface depend only on their immediate neighbors. In symbolic music corpora, for example, researchers often divide the corpus into contiguous sequences of n events (called n-grams) for the purposes of pattern discovery, classification, similarity estimation, and prediction. And yet since much of the world's music is hierarchically organized such that certain events are more stable (or prominent) than others, non-contiguous events often serve as focal points in the sequence. Thus, this chapter considers how we might adapt skip-grams, which include sub-sequences in an n-gram distribution if their constituent members occur within a certain number of skips, to address questions related to the analysis of tonal harmony in a corpus of Haydn string quartets.