Pulse waveform classification using ERP-based difference-weighted KNN classifier

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

5 Citations (Scopus)

Abstract

Although the great progress in sensor and signal processing techniques have provided effective tools for quantitative research into traditional Chinese pulse diagnosis, the automatic classification of pulse waveform is remained a difficult problem. In order to address this issue, we propose a novel edit distance with real penalty-based k-nearest neighbor classifier by referring to recent progress in time series matching and KNN classifier. Taking advantage of the metric property of ERP, we develop an ERP-induced inner product operator and then embed it into difference-weighted KNN classifier. Experimental results show that the proposed classifier is more accurate than comparable pulse waveform classification approaches.

Original languageEnglish
Title of host publicationMedical Biometrics - Second International Conference, ICMB 2010, Proceedings
Pages191-200
Number of pages10
DOIs
Publication statusPublished - 2010
Event2nd International Conference on Medical Biometrics, ICMB 2010 - Hong Kong, China
Duration: 28 Jun 201030 Jun 2010

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6165 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference2nd International Conference on Medical Biometrics, ICMB 2010
Country/TerritoryChina
CityHong Kong
Period28/06/1030/06/10

Keywords

  • edit distance with real penalty (ERP)
  • k-nearest neighbor (KNN)
  • pulse waveform
  • time series classification

Fingerprint

Dive into the research topics of 'Pulse waveform classification using ERP-based difference-weighted KNN classifier'. Together they form a unique fingerprint.

Cite this