Approximate Nearest Neighbour Search with the Fukunaga and Narendra Algorithm and Its Application to Chromosome Classification
The nearest neighbour (NN) rule is widely used in pattern recognition tasks due to its simplicity and its good behaviour. Many fast NN search algorithms have been developed during last years. However, in some classification tasks an exact NN search is too slow, and a way to quicken the search is required. To face these tasks it is possible to use approximate NN search, which usually increases error rates but highly reduces search time.
In this work we propose using approximate NN search with an algorithm suitable for general metric spaces, the Fukunaga and Narendra algorithm, and its application to chromosome recognition. Also, to compensate the increasing in error rates that approximate search produces, we propose to use a recently proposed framework to classify using k neighbours that are not always the k nearest neighbours. This framework improves NN classification rates without extra time cost.
KeywordsApproximate Nearest Neighbour Pattern Recognition Chromosome Recognition
- 2.Brin, S.: Near Neighbor Search in Large Metric Spaces. In: Proceedings of the 21st VLDB Conference, pp. 574–584 (1995)Google Scholar
- 7.Granum, E., Thomason, M.G., Gregor, J.: On the use of automatically inferred Markov networks for chromosome analysis. In: Lundsteen, C., Piper, J. (eds.) Automation of Cytogenetics, pp. 233–251. Springer, Heidelberg (1989)Google Scholar
- 13.Yianilos, P.N.: Data Structures and Algorithms for Nearest Neighbor Search in General Metric Spaces. In: ACM-SIAM Symposium on Discrete Algorithms, pp. 311–321 (1993)Google Scholar