The LVQ algorithm is a common method which allows a set of reference vectors for a distance classifier to adapt to a given training set. We have developed a similar learning algorithm, LVQ-MM, which manipulates hyperellipsoidal cluster boundaries as opposed to reference vectors. Regions of the input feature space are first enclosed by ellipsoidal decision boundaries, and then these boundaries are iteratively modified to reduce classification error. Results obtained by classifying the Iris data set are provided.
P.M. Kelly, D.R. Hush, and J.M. White. An adaptive algorithm for modifying hyperellipsoidal decision surfaces. In Proceedings of the International Joint Conference on Neural Networks, Vol. 4, pages 196-201, 1992. Los Alamos National Laboratory Technical Report LA-UR-92-1261. [ Abstract | PostScript (155 KB) | PDF (135 KB) ]
