A self organizing feature map based on the fisher discriminant
Document typeConference report
Rights accessOpen Access
In this paper we propose a modification of the Self Organising map, where we is train the units using a transformation related to the Fisher Discriminant, in order to maximise the discrimination between units. The algorithm integrates the recursion for updating the Self Organising Maps with the recursion for updating the discrimination matrix.
CitationMonte, E. A self organizing feature map based on the fisher discriminant. A: International Conference on Spoken Language Processing. "ICSLP 94: International Conference on Spoken Language Processing: Pacific Convention Plaza Yokohama (PACIFICO), Yokohama, Japan: September 18-22, 1994". Yokohama: 1994.