C-Pruner: An improved instance pruning algorithm

  • Ke Ping Zhao*
  • , Shui Geng Zhou
  • , Ji Hong Guan
  • , Ao Ying Zhou
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

26 Scopus citations

Abstract

Instance-based learning faces the problem of deciding which instances could be discarded in order to save computation and storage costs. For large instance bases classifier suffers from large memory requirements and slow response. And present noisy instances may deteriorate the classification accuracy. This paper analyzes the strength and weakness of some of the existing algorithms for instance pruning, and propose an improved method C-Pruner. Experiments over real-world datasets verify C-Pruner's superior to the existing methods in classification accuracy.

Original languageEnglish
Title of host publicationInternational Conference on Machine Learning and Cybernetics
Pages94-99
Number of pages6
StatePublished - 2003
Externally publishedYes
Event2003 International Conference on Machine Learning and Cybernetics - Xi'an, China
Duration: 2 Nov 20035 Nov 2003

Publication series

NameInternational Conference on Machine Learning and Cybernetics
Volume1

Conference

Conference2003 International Conference on Machine Learning and Cybernetics
Country/TerritoryChina
CityXi'an
Period2/11/035/11/03

Keywords

  • Instance Pruning
  • Instance-based learning
  • KNN classification

Fingerprint

Dive into the research topics of 'C-Pruner: An improved instance pruning algorithm'. Together they form a unique fingerprint.

Cite this