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

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. © 2010 Springer-Verlag.