Characterization of climatological time series using autoencoders

Reynaldo Alfonte Zapana, Cristian José Lopez Del Alamo, Jan Franco Llerena Quenaya, Ana Maria Cuadros Valdivia

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

1 Scopus citations

Abstract

Common problems in climatological time series data are high dimensionality, correlation between the sequential values and noise due to calibration of meteorological stations influencing dramatically in the quality of clustering, classification, climate pattern finding and data processing. One way to deal with this problem is through feature extraction technique. In order to extract features from large climatological time series data, we propose a feature extraction method based on autoencoder neural network (AUTOE). As a first step, time series is standardized. Then, different architectures of autoencoder is applied on it to reduce dimensionality. Finally, k-means clustering algorithm are used to evaluate them through quality measures. As a result, autoencoder performs well and is competitive with other feature extraction techniques over Synthetic Control Chart Time Series.

Original languageEnglish
Title of host publication2017 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2017 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1-6
Number of pages6
ISBN (Electronic)9781538637340
DOIs
StatePublished - 7 Feb 2018
Event2017 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2017 - Arequipa, Peru
Duration: 8 Nov 201710 Nov 2017

Publication series

Name2017 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2017 - Proceedings
Volume2017-November

Conference

Conference2017 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2017
Country/TerritoryPeru
CityArequipa
Period8/11/1710/11/17

Bibliographical note

Publisher Copyright:
© 2017 IEEE.

Keywords

  • autoencoder
  • Dimensionality reduction
  • time series

Fingerprint

Dive into the research topics of 'Characterization of climatological time series using autoencoders'. Together they form a unique fingerprint.

Cite this