A Co-training approach for time series prediction with missing data

Tawfik A. Mohamed, Neamat El Gayar, Amir F. Atiya

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

2 Citations (Scopus)

Abstract

In this paper we consider the problem of missing data in time series analysis. We propose a semi-supervised co-training method to handle the problem of missing data. We transform the time series data to set of labeled and unlabeled data. Different predictors are used to predict the unlabelled data and the most confident labeled patterns are used to retrain the predictors further to and enhance the overall prediction accuracy. By labeling the unknown patterns the missing data is compensated for. Experiments were conducted on different time series data and with varying percentage of missing data using a uniform distributions We used KNN base predictors and Fuzzy Inductive Reasoning (FIR) base predictors and compared their performance using different confidence measures. Results reveal the effectiveness of the co-training method to compensate for the missing values and to improve prediction. The FIR model together with the "similarity" confidence measures obtained in most cases the best results in our study.

Original languageEnglish
Title of host publicationMultiple Classifier Systems. MCS 2007
PublisherSpringer
Pages93-102
Number of pages10
ISBN (Electronic)9783540725237
ISBN (Print)9783540724810
DOIs
Publication statusPublished - 2007
Event7th International Workshop on Multiple Classifier Systems 2007 - Prague, Czech Republic
Duration: 23 May 200725 May 2007

Publication series

NameLecture Notes in Computer Science
Volume4472
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference7th International Workshop on Multiple Classifier Systems 2007
Abbreviated titleMCS 2007
CountryCzech Republic
CityPrague
Period23/05/0725/05/07

Keywords

  • Co-training
  • Ensemble prediction
  • Fuzzy inductive reasoning
  • K-nearest neighbor
  • Missing data
  • Semi-supervised learning
  • Time series prediction

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'A Co-training approach for time series prediction with missing data'. Together they form a unique fingerprint.

Cite this