Issue No. 04 - April (1989 vol. 11)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/34.19040
The Parzen density estimate is known to be an effective tool for estimating the Bayes error, given a set of training samples from the class distributions. An algorithm is developed to select a given number of representative samples whose Parzen density estimate closely matches that of the entire sample set. Using this reduced representative set, a piecewise quadratic classifier which provides nearly optimal performance is designed.<
pattern recognition, Bayes methods, error analysis, estimation theory, piecewise quadratic classifier, pattern recognition, Parzen classifier, Parzen density estimate, Bayes error, representative samples, Gaussian distribution, Error analysis, Kernel, Shape, Millimeter wave radar, Pattern recognition, Covariance matrix, Probability distribution, Design optimization, Algorithm design and analysis
"The reduced Parzen classifier," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 11, no. , pp. 423,424,425, 1989.