Time series labeling algorithms based on the K-nearest neighbors' frequencies

Nasibov E., Peker S.

EXPERT SYSTEMS WITH APPLICATIONS, vol.38, no.5, pp.5028-5035, 2011 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 38 Issue: 5
  • Publication Date: 2011
  • Doi Number: 10.1016/j.eswa.2010.09.147
  • Journal Name: EXPERT SYSTEMS WITH APPLICATIONS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.5028-5035
  • Keywords: Time series, Clustering, FCM, K-nearest neighbor, Bispectral index, CLUSTER VALIDITY, MODEL
  • Dokuz Eylül University Affiliated: Yes

Abstract

In the current paper, time series labeling task is analyzed and some solution algorithms are presented. In these algorithms, fuzzy c-means clustering, which is one of the unsupervised learning methods, is used to obtain the labels of the time series. Then K-nearest neighborhood (KNN) rule is performed on the labels to obtain more relevant smooth intervals.