Issue No. 02 - February (1968 vol. 17)
W.G. Wee , IEEE
Abstract?A more general adaptive procedure for determining linear or piecewise linear discriminant functions for multiclass pattern classification is proposed. The adaptive procedure is a many-pattern or group-pattern adaptation. The training sequence consists of groups of vectors in matrix form instead of single vectors. The convergence proof shows that this procedure terminates in a finite number of adaptions if the solution exists. A necessary and suffcient condition is developed for testing the linear separability of each subset of (d + 1) samples. Furthermore, the proposed procedure can be implemented with the addition of only a little complexity to existing systems. Computer simulations indecate satisfactory results.
Index terms?Adaptive procedure, discriminant function, linear separability, multiclass, pattern recognition, single-pattern adaptation, total-pattern adaptation.
K.S. Fu, W.G. Wee, "An Adaptive Procedure for Multiclass Pattern Classification", IEEE Transactions on Computers, vol. 17, no. , pp. 178-182, February 1968, doi:10.1109/TC.1968.227406