A novel ensemble method for time series classification

Sami M. Halawani, Ibrahim A. Albidewi, Amir Ahmad

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)


This paper explores the issue of input randomization in decision tree ensembles for time series classification. We suggest an unsupervised discretization method to create diverse discretized datasets.We introduce a novel ensemble method, in which each decision tree is trained on one dataset from the pool of different discretized datasets created by the proposed discretization method. As the discretized data has a small number of boundaries the decision tree trained on it is forced to learn on these boundaries. Different decision trees trained on datasets having different discretization boundaries are diverse. The proposed ensembles are simple but quite accurate. We study the performance of the proposed ensembles against the other popular ensemble techniques. The proposed ensemble method matches or outperforms Bagging, and is competitive with Adaboost.M1 and Random Forests.

Original languageEnglish
Title of host publicationComputer Networks and Intelligent Computing - 5th International Conference on Information Processing, ICIP 2011, Proceedings
Number of pages6
Publication statusPublished - 2011
Externally publishedYes
Event5th International Conference on Information Processing, ICIP 2011 - Bangalore, India
Duration: Aug 5 2011Aug 7 2011

Publication series

NameCommunications in Computer and Information Science
Volume157 CCIS
ISSN (Print)1865-0929


Conference5th International Conference on Information Processing, ICIP 2011


  • Classification
  • Decision trees
  • Ensembles
  • Time series

ASJC Scopus subject areas

  • General Computer Science
  • General Mathematics


Dive into the research topics of 'A novel ensemble method for time series classification'. Together they form a unique fingerprint.

Cite this