A hierarchical self-organising map model for sequence recognition

This raper presents art analysis of an original hierarchical neural model o n a complex sequence - che complete sixteenth four-part fugue in G minor of the Well-Tempered Clavier (vol. I) of J. S. Bach. The model makes an effec tive use of context information, through in hierarchical topology and its e mbedded time integrators, and that enables it to keep a very good account o f patterns. The model peri;,rms sequence classification and discrimination efficiently. It has application in domains which require pattern recognitio n, or particularly, which demand recognising either a set of sequences of v ectors in time, or sub-sequences into a unique and large sequence of vector s in time.