This Article 
 Bibliographic References 
 Add to: 
Fast Nearest-Neighbor Search in Dissimilarity Spaces
September 1993 (vol. 15 no. 9)
pp. 957-962

A fast nearest-neighbor algorithm is presented. It works in general spaces in which the known cell techniques cannot be implemented for various reasons, such as the absence of coordinate structure or high dimensionality. The central idea has already appeared several times in the literature with extensive computer simulation results. An exact probabilistic analysis of this family of algorithms that proves its O(1) asymptotic average complexity measured in the number of dissimilarity calculations is presented.

[1] D. Dobkin and R. J. Lipton, "Multidimensional searching problems,"SIAM J. Comput., vol. 5, no. 2, pp. 181-186, June 1976.
[2] J. H. Friedman, J. L. Bentley, and R. A. Finkel, "An algorithm for finding best matches in logarithmic expected time,"ACM Trans. Math. Software, vol. 3, no. 3, pp. 209-226, Sept. 1977.
[3] J. L. Bentley, B. W. Weide, and A. C. Yao, "Optimal expected time algorithms for closest point problems,"ACM Trans. Math. Software, vol. 6, pp. 563-580, 1980.
[4] I. K. Sethi, "A fast algorithm for recognizing nearest neighbors,"IEEE Trans. Syst. Man Cyber., vol. SMC-11, pp. 245-248, Mar. 1981.
[5] E. Vidal, "An algorithm for finding nearest neighbors in (approximately) constant average time,"Patt. Recogn. Lett., vol. 4, no. 3, pp. 145-157, July 1986.
[6] A. Faragó, T. Linder, G. Lugosi, and T. Pikler, "On the algorithmic problems of the nearest neighbor method,"Hiradástechnika (Telecommunication), vol. XXXIX, no. 8, 1988; in Hungarian.
[7] K. Motoishi and T. Misumi, "Fast vector quantization algorithm by using an adaptive search technique," presented atIEEE Int. Symp. Inform. Theory(San Diego, CA), Jan. 14-19, 1990.
[8] T. Linder and G. Lugosi, "Classification with a low complexity nearest neighbor algorithm," presented atIEEE Int. Symp. Inform. Theory(San Diego, CA), Jan. 14-19, 1990.
[9] L. Devroye,Lecture Notes on Bucket Algorithms. Boston: Birkhäuser, 1986.
[10] E. Vidal, H. Rulot, F. Casacuberta, and J. Benedi, "On the use of metric-space search algorithm (AESA) for fast DTW-based recognition of isolated words,"IEEE Trans. Acoust. Speech. Signal Processing, vol. ASSP-36, pp. 651-660, 1988.
[11] R. L. Wheeden and A. Z. Zygmund,Measure and Integral. New York: Marcel Dekker, 1977.
[12] C. Berge,Graphs and Hypergraphs. Amsterdam: North Holland, 1973.
[13] E. Vidal and M. J. Lloret, "Recent results on the application of a metric-space seaeh algorithm (AESA) to multispeaker data," inRecent Advances in Speech Understanding and Dialog Systems(H. Niemann, Ed.). New York: Springer Verlag, 1988.

Index Terms:
high-dimensional spaces; nearest-neighbor search; dissimilarity spaces; fast nearest-neighbor algorithm; exact probabilistic analysis; O(1) asymptotic average complexity; dissimilarity calculations; computational complexity; pattern recognition; search problems
A. Faragó, T. Linder, G. Lugosi, "Fast Nearest-Neighbor Search in Dissimilarity Spaces," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 15, no. 9, pp. 957-962, Sept. 1993, doi:10.1109/34.232083
Usage of this product signifies your acceptance of the Terms of Use.