Showing papers on "Adaptive algorithm published in 1967"
IBM1
TL;DR: The algorithm discussed offers two advantages for pattern recognition applications: the number of patterns which must be labeled with class identification is reduced, and the adaptive system can follow changes in the class distributions over time, due to data fluctuation or hardware degradation.
Abstract: An unsupervised or nonsupervised adaptive algorithm for linear decision boundaries is applied to two pattern recognition problems: the classification of spoken words, and the classification of hand-printed characters. The term unsupervised indicates that the class identification of the input patterns is not continuously available to the adaptive system. The algorithm discussed offers two advantages for pattern recognition applications. First, the number of patterns which must be labeled with class identification is reduced. Second, the adaptive system can follow changes in the class distributions over time, due to data fluctuation or hardware degradation. These advantages are demonstrated for each of the two applications.
4 citations
IBM1
TL;DR: Adaptively derived linear decision boundaries are used in the recognition of typewritten numeric characters from a variety of fonts by allowing as many quantization levels as required by the adaptive algorithm.
Abstract: Adaptively derived linear decision boundaries are used in the recognition of typewritten numeric characters from a variety of fonts. During the adaption process, the weights are allowed to have as many quantization levels as required by the adaptive algorithm. Once these weights and their recognition rate on a test sample are determined, the number of quantization levels is successively reduced, and the effect on recognition rate is tabulated. System performance is maintained surprisingly well as the number of weight levels is reduced.