Reducing the Computational Cost of Computing Approximated Median Strings

Authors:
Carlos D. Martínez-Hinarejos;Alfonso Juan;Francisco Casacuberta;Ramón Alberto Mollineda
Affiliations:
-;-;-;-
Venue:
Proceedings of the Joint IAPR International Workshop on Structural, Syntactic, and Statistical Pattern Recognition
Year:
2002

Citing 5
Cited 3

Topology of strings: median string is NP-complete

Theoretical Computer Science
Fast Computation of Normalized Edit Distances

IEEE Transactions on Pattern Analysis and Machine Intelligence
Median strings for k-nearest neighbour classification

Pattern Recognition Letters
Fast Median Search in Metric Spaces

SSPR '98/SPR '98 Proceedings of the Joint IAPR International Workshops on Advances in Pattern Recognition
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)

A Stochastic Approach to Median String Computation

SSPR & SPR '08 Proceedings of the 2008 Joint IAPR International Workshop on Structural, Syntactic, and Statistical Pattern Recognition
Improved background modeling for real-time spatio-temporal non-parametric moving object detection strategies

Image and Vision Computing
A new iterative algorithm for computing a quality approximate median of strings based on edit operations

Pattern Recognition Letters

Quantified Score

Hi-index	0.00

Visualization

Abstract

The k-Nearest Neighbour (k-NN) rule is one of the most popular techniques in Pattern Recognition. This technique requires good prototypes in order to achieve good results with a reasonable computational cost. When objects are represented by strings, the Median String of a set of strings could be the best prototype for representing the whole set (i.e., the class of the objects). However, obtaining the Median String is an NP-Hard problem, and only approximations to the median string can be computed with a reasonable computational cost. Although proposed algorithms to obtain approximations to Median String are polynomial, their computational cost is quite high (cubic order), and obtaining the prototypes is very costly. In this work, we propose several techniques in order to reduce this computational cost without degrading the classification performance by the Nearest Neighbour rule.