A Nonparametric Two-Dimensional Display for Classification

A two-dimensional display whose coordinates are related to the distance to the kth-nearest neighbor of each class is presented. Applications of the display to minimum error, minimum cost, minimax, and Neyman-Pearson type classifier designs are given. The display is shown to present risk information in a manner that easily allows the specification of reject regions. Two methods of error estimation using the display, an error counting technique and a risk averaging method, are detailed. It is shown that the classifiers that result are generalizations of the standard k-NN majority vote classifier. As a result of the properties of the display, classifiers can be readily evaluated and modified. In addition, a condensing algorithm that preserves the nearest neighbor error count of any preclassified data set is described. The display is used to graphically illustrate the distance relationships that are central to the algorithm.

[1]  Jack Koplowitz,et al.  The weighted nearest neighbor rule for class dependent sample sizes (Corresp.) , 1979, IEEE Trans. Inf. Theory.

[2]  Marshall W. Anderson,et al.  A distribution-free discrimination procedure based on clustering , 1970, IEEE Trans. Inf. Theory.

[3]  C. G. Hilborn,et al.  The Condensed Nearest Neighbor Rule , 1967 .

[4]  Theodosios Pavlidis,et al.  Computer Recognition of Handwritten Numerals by Polygonal Approximations , 1975, IEEE Transactions on Systems, Man, and Cybernetics.

[5]  M Berthod,et al.  Learning in syntactic recognition of symbols drawn on a graphic tablet , 1979 .

[6]  Peter E. Hart,et al.  The condensed nearest neighbor rule (Corresp.) , 1968, IEEE Trans. Inf. Theory.

[7]  Keinosuke Fukunaga,et al.  A Two-Dimensional Display for the Classification of Multivariate Data , 1971, IEEE Transactions on Computers.

[8]  John W. Sammon,et al.  A Nonlinear Mapping for Data Structure Analysis , 1969, IEEE Transactions on Computers.

[9]  GUY W. BEAKLEY,et al.  Distribution-Free Pattern Verification Using Statistically Equivalent Blocks , 1972, IEEE Transactions on Computers.

[10]  E. Oja,et al.  On-line recognition of handwritten characters , 2004 .

[11]  Larry D. Hostetler,et al.  k-nearest-neighbor Bayes-risk estimation , 1975, IEEE Trans. Inf. Theory.

[12]  John W. Sammon,et al.  An Optimal Discriminant Plane , 1970, IEEE Transactions on Computers.

[13]  Keinosuke Fukunaga,et al.  Estimation of Classification Error , 1970, IEEE Transactions on Computers.

[14]  Ivan Tomek,et al.  A Generalization of the k-NN Rule , 1976, IEEE Transactions on Systems, Man, and Cybernetics.

[15]  Robert Dunham Short.  The feature extraction problem for statistical pattern recognition. , 1978 .

[16]  M. Thomason.  Interactive Pattern Recognition , 1981, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[17]  Waren Teitelman,et al.  Real time recognition of hand-drawn characters , 1964, AFIPS '64 (Fall, part I).

[18]  G. Krishna,et al.  Agglomerative clustering using the concept of mutual nearest neighbourhood , 1978, Pattern Recognit..

[19]  Richard C. T. Lee,et al.  A Triangulation Method for the Sequential Mapping of Points from N-Space to Two-Space , 1977, IEEE Transactions on Computers.

[20]  C. K. Chow,et al.  On optimum recognition error and reject tradeoff , 1970, IEEE Trans. Inf. Theory.

[21]  M. Gessaman,et al.  Nonparametric Discrimination Using Tolerance Regions , 1968 .

[22]  Roger W. Ehrich,et al.  Experiments in the Contextual Recognition of Cursive Script , 1975, IEEE Transactions on Computers.