Towards decision rule based on closer symmetric neighborhood

• Autorzy: Grabowski S.
• Języki publikacji: EN

Abstrakty

EN

The relatively new k Nearest Centroid Neighbor (k-NCN) decision rule uses an interesting concept of surrounding neighborhood, that is, such a neighborhood which takes into account not only the proximity of neighbors but also their spatial location. In the paper we propose a new decision rule, called k Near Surrounding Neighbors (k-NSN), which "improves" the neighborhood used in k-NCN with respect to both mentioned aspect. We tested several methods, k-NN included, each in a multidecision and in two binary decomposition schemes, on a few UCI datasets and a large ferrite core dataset, to show attractiveness of the presented concept in applications where the prediction accuracy is of utmost importance.

Słowa kluczowe

EN

pattern recognition; k Nearest Centroid Neighbors; k Near Surrounding Neigbors; pairwise decomposition

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2003
• Tom: Vol. 23, no. 3
• Strony: 39--46
• Opis fizyczny: Bibliogr. 19 poz., rys., tab., wykr.

Twórcy

autor

Grabowski S.

• Technical University of Łódź, Poland,

Bibliografia

• [1] Fukunaga K.: Introduction to statistical pattern recognition. Academic Press: San Diego, CA, 1990.
• [2] Wilson D.L.: Asymptotic properties of nearest neighbour rules using edited data. IEEE Transactions on Systems, Man and Cybernetics, July 1972, SMC-2, 3, 408-421.
• [3] Tomek I.: An experiment with the edited nearest-neighbour rule. IEEE Trans, on Systems, Man and Cybernetics, 1976, SMC-6, 6, 448-452.
• [4] Dudani S.A.: The distance weighted k-nearest neighbour rule. IEEE Transactions on Systems, Man and Cybernetics, 1976, SMC-6, 325-327.
• [5] Jóźwik A.: A learning scheme for a fuzzy k-NN rule. Pattern Recognition Letters, 1983, 1, 287-289.
• [6] Short R.D., Fukunaga K.: The optimal distance measure for nearest neighbor classification. IEEE Transactions on Information Theory, 1981,27, 622-627.
• [7] Chaudhuri B.B.: A new definition of neighbourhood of a point in multi-dimensional space. Pattern Recognition Letters, 1996, 17, 11-17.
• [8] Sanchez J.S., Pla F., Ferri F.J.: On the use of neighbourhood-based non-parametric classifiers. Pattern Recognition Letters, 1997, 18, 1179-1186.
• [9] Sanchez J.S., Pla F„ Ferri F.J.: Improving the k-NCN classification rule through heuristic modifications. Pattern Recognition Letters, 1998, 19, 1165-1170.
• [10] Jóźwik A., Vernazza G.: Recognition of leucocytes by a parallel k-NN classifiers. Lecture Notes of ICB Seminar, 138-153, Warsaw 1988.
• [11] Dietterich T.G., Bakiri G.: Error-correcting output codes: A general method of improving multiclass inductive learning programs. Proceedings of the Ninth National Conference on Artificial Intelligence (AAA-91), Anaheim, CA: AAAI Press, 1991.
• [12] Bay S.D.: Combining nearest neighbor classifiers through multiple feature subsets. Proceedings of the Fifteenth International Conference on Machine Learning. Madison, WI, 1998, 37-45.
• [13] Kong E.B., Dietterich T.G.: Error-correcting output coding corrects bias and variance. Proceedings of the Twelfth International Conference on Machine Learning, 313-321, Tahoe City, CA. Morgan Kaufmann, 1995.
• [14] Moreira M„ Mayoraz E.: Improved pairwise coupling classification with correcting classifiers. Proceedings of the Tenth European Conference on Machine Learning, Chemnitz, Germany 1998 160-171.
• [15] Masuli F., Valentini G.: Comparing decomposition methods for classification. In: R.J. Howlett and L.C. Jain, editors, KES’2000, Fourth International Conference on Knowledge-Based Intelligent Engineering Systems & Allied Technologies, Piscataway, NJ, 2000, 788-791.
• [16] Skalak D.B.: Prototype and feature selection by sampling and random mutation hill climbing algorithms. In Machine Learning: Proceedings of the Eleventh International Conference, Morgan Kaufmann, 1994, 293-301.
• [17] Jóźwik A., Chmielewski L., Skłodowski M„ Cudny W.: A parallel net of (1-NN, k-NN) classifiers for optical inspection of surface defects in ferrites. Machine Graphics and Vision 1998 7 (1-2) 99-112.
• [18] Merz C„ Murphy P.M.: UCI repository of machine learning databases, [http://www.ics.uci.edu/mleam/MLRepository.html], 1996.
• [19] Jóźwik A., Chmielewski L., Cudny W., Skłodowski M.: A 1-NN preclassifier for fuzzy k-NN rule. Proceedings of the Thirteenth International Conference on Pattern Recognition, vol. IV, track D, Parallel and Connectionist Systems, 1996, 234-238, Vienna, Austria, August 25-29.