editorial applications of computational intelligence in...

EditorialApplications of Computational Intelligence in Time Series

Francisco Martínez-Álvarez,1 Alicia Troncoso,1 Jorge Reyes,2

María Martínez-Ballesteros,3 and José C. Riquelme3

1Division of Computer Science, Pablo de Olavide University, 41013 Seville, Spain2NT2 Labs, Santiago, Chile3Department of Computer Science, University of Seville, Sevilla, Spain

Correspondence should be addressed to Francisco Martınez-Alvarez; [email protected]

Received 26 April 2017; Accepted 26 April 2017; Published 22 May 2017

Copyright © 2017 Francisco Martınez-Alvarez et al. �is is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

�e prediction of the future has fascinated the human beingsince its early existence. Actually, many of these e�orts canbe noticed in everyday events such as energy management,telecommunications, pollution, bioinformatics, and seismol-ogy and, obviously, in neuroscience. Accurate predictionsare essential in economic activities as remarkable forecastingerrors in certain areas may involve large loss of money.

In this context, the successful analysis of temporal datahas been a challenging task for many researchers duringthe last decades and, indeed, it is dicult to gure out anyscientic branch with no time-dependent variables.

Computational intelligence is known for including pow-erful techniques such as articial neural networks, fuzzysystems, evolutionary computation, learning theory, or prob-abilistic methods. �us, this special issue has been focusedon the application of such techniques to time series. Inparticular, the goal was sharing recent advances in timeseries analysis and providing an interesting opportunity topresent and discuss the latest practical advances in real-worldapplications.

Rigorous and extensive review processes have been car-ried out. Papers selected for this special issue present newndings and insights in the eld of time series forecasting.A broad range of topics has been discussed, especially in thefollowing areas: nances, tourism, feed grain demand, hazeepisodes, stock price, or data storage.

Genetic algorithms have been developedwith binary cod-ing to analyze high-speed trading research using price data ofstocks on themicroscopic level.�is problem is certainly new

and unexplored from computational intelligence techniques.Reported results show that the system is able to improve theaccuracy for price movement forecasting, thus encouragingconducting research in this direction.

Tourism demand forecasting has also been addressed bymeans of seasonal trend autoregressive integrated movingaverages with dendritic neural networks. Data from Japanare used to test the predictive performance of the proposedmodel. �e model generates short-term predictions, a�erapplying SARIMA models to exclude the long-term lineartrend. �is study mixes linear and nonlinear models andsuggests that further analysis in the combination of suchtechniques is desirable.

Likewise, a new irregular sampling estimation methodto extract the main trend of the time series has beenproposed. To achieve this, rst, the Kalman lter is used toremove dirty data. Second, the cubic spline interpolation andaverage method are used to reconstruct the main trend. �eproposed approach has been applied to storage volume seriesof Internet data center. Results are quite promising.

A novel hybrid articial intelligent system has beenproposed to forecast stock price index trend. In particular,articial neural networks combined with genetic algorithmshave been applied to data from�ailand’s SET50 index trend,from years 2009 to 2014. Multiple features and di�erent timespans have been considered. Comparisons to other methodsconrm the success of this novel methodology.

�e long-term prediction of feed grain demand issuehas been extensively discussed. A multivariate regression

HindawiComputational Intelligence and NeuroscienceVolume 2017, Article ID 9361749, 2 pageshttps://doi.org/10.1155/2017/9361749

https://doi.org/10.1155/2017/9361749

2 Computational Intelligence and Neuroscience

model along with a dynamic forecasting model has beenintroduced. Firstly, the correlation between the demand andits in�uence factors are studied and, secondly, changes intrend in factors a�ecting the demand are forecasted. Reportedresults corroborate the e�ectiveness of the methodology.

A novel long-term prediction model for Beijing hazeepisodes has been introduced. �e authors have built adynamic structural measurement model of daily haze incre-ment and have reduced the model to a vector autoregressivemodel. Such model performs satisfactorily on next day’s airquality index forecasting, reaching in many cases an accuraterate close to 90%. �erefore, sudden haze burst could bepredicted with this method.

Acknowledgments

Finally, we would like to thank all the authors for theirexcellent work and contributions to this special issue. Wewould also like to express our gratitude to all the reviewersfor their thorough revisions and patience in assisting us.

Francisco Martınez-AlvarezAlicia Troncoso

Jorge ReyesMarıa Martınez-Ballesteros

Jose C. Riquelme

Submit your manuscripts athttps://www.hindawi.com

Computer Games Technology

International Journal of

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


Distributed Sensor Networks


Advances in

FuzzySystems

Hindawi Publishing Corporationhttp://www.hindawi.com

Volume 2014


ReconfigurableComputing

Hindawi Publishing Corporation http://www.hindawi.com Volume 2014


Applied Computational Intelligence and Soft Computing

Advances in

Artificial Intelligence


Advances inSoftware EngineeringHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


Electrical and Computer Engineering

Journal of

Hindawi Publishing Corporation

http://www.hindawi.com Volume 2014

Advances in

Multimedia


Biomedical Imaging


Advances in


RoboticsJournal of



Computational Intelligence and Neuroscience

Industrial EngineeringJournal of


Modelling & Simulation in EngineeringHindawi Publishing Corporation http://www.hindawi.com Volume 2014

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014


Human-ComputerInteraction

Advances in

Computer EngineeringAdvances in


editorial applications of computational intelligence in...

Documents