Fast Design of Reduced-Complexity Nearest-Neighbor Classifiers Using Triangular Inequality

Authors:
Eei-Wan Lee;Soo-Ik Chae
Affiliations:
Seoul National Univ., Seoul, Korea;Seoul National Univ., Seoul, Korea
Venue:
IEEE Transactions on Pattern Analysis and Machine Intelligence
Year:
1998

Citing 0
Cited 5

A modification of the LAESA algorithm for approximated k-NN classification

Pattern Recognition Letters
Data mining tasks and methods: Classification: nearest-neighbor approaches

Handbook of data mining and knowledge discovery
Fast and versatile algorithm for nearest neighbor search based on a lower bound tree

Pattern Recognition
New Algorithms for Efficient High-Dimensional Nonparametric Classification

The Journal of Machine Learning Research
A fast fuzzy $K$-nearest neighbour algorithm for pattern classification

Intelligent Data Analysis

Quantified Score

Hi-index	0.14

Visualization

Abstract

In this paper, we propose a method of designing a reduced complexity nearest-neighbor (RCNN) classifier with near-minimal computational complexity from a given nearest-neighbor classifier that has high input dimensionality and a large number of class vectors. We applied our method to the classification problem of handwritten numerals in the NIST database. If the complexity of the RCNN classifier is normalized to that of the given classifier, the complexity of the derived classifier is 62 percent, 2 percent higher than that of the optimal classifier. This was found using the exhaustive search.