-
The first mission of the Cantorinus project is to gather the ΔΩΔΕΚΑΧΟΡΔΟΝ (Dodecachordon) dataset. The dataset consists of all monophonic, and polyphonic compositions listed in Glarean's ΔΩΔΕΚΑΧΟΡΔΟΝ.
-
Next the Cantorinus project is to assemble rule based mode finding algorithms.
- The first set of heuristic algorithms are adapted from Peter Schubert's Modal Counterpoint, Renaissance Style book.
- The second set of heuristic algorithms are based on Glarean's ΔΩΔΕΚΑΧΟΡΔΟΝ treatise from 1547.
- Once both tasks have been completed different types of machine learning algorithms will be applied to the dataset, and a test set.
- Supervised/guided machine learning
- Semi-supervised machine learning
- Unsupervised machine learning