Skip to main content
SpringerLink
Log in

FRASel: a consensus of feature ranking methods for time series modelling

  • Methodologies and Application
  • Published:
Soft Computing Aims and scope Submit manuscript

Abstract

The main objective of “time series analysis” is to discover the underlying structure of the time series, and thus, become able to forecast its “future values”. This process makes it possible to predict, control or simulate variables. Most of the time series modelling procedures try to forecast future values from lagged ones. Thus, the selection of the relevant lagged values to be used is a key step. In this paper, a new consensus method for the selection of relevant lagged values of a time series is introduced: feature ranking aggregated selection (FRASel). The main contribution of this feature selection method is the definition of a consensus decision making mechanism based on aggregation and expressed as a simple rule. In FRASel, the selected subset of lagged values is decided by the application of an aggregation criterion to the results of different flavours of feature ranking methods, applied from different approaches. A thorough empirical analysis is carried out to assess the performance of FRASel. The statistical significance of the experimental results is also analysed through the application of non-parametric statistical tests.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Notes

  1. Individual comparisons available at http://dicits.ugr.es/papers/frasel.

  2. FRASel synthetic time series files (FRASelserie*.dat) available at http://sci2s.ugr.es/keel/datasets.

  3. NNGC1 time series files available at http://www.neural-forecasting-competition.com.

  4. Different seeds for pseudo-random number generators.

  5. Complete experimental results available at http://dicits.ugr.es/papers/frasel..

  6. In FRASel, N is the number of features considered as relevant candidates in the aggregation phase. In NIMFS, K is the the final subset size. The other methods—wrappers and filters—obtain the subsets without a parameter

  7. Statistical tests are computed at a 95 % confidence level. H 0, indicates that the two feature selection strategies provide the same subsets size. H 1 reflects that the amount of features selected by FRASel is smaller.

  8. H 0 indicates that the two samples have the same error in the test phase. H 1 reflects that models built with FRASel selections outperforms models built with the selections provided by the alternative method.

References

  • Amjady N, Keynia F, Zareipour H (2010) Short-term load forecast of microgrids by a new bilevel prediction strategy. IEEE Transactions Smart Grid 1(3):286–294

    Article  Google Scholar 

  • Araúzo-Azofra A, Benítez JM, Castro JL (2008) Consistency measures for feature selection. J Intell Inf Syst 30:3:273–292

    Article  Google Scholar 

  • Araúzo-Azofra A, Aznarte JL, Benítez JM (2011) Empirical study of feature selection methods based on individual feature evaluation for classification problems. Expert Syst Appl 38(7):8170–8177

    Article  Google Scholar 

  • Ben-Bassat M (1982) Pattern Recognition and Reduction of Dimensionality. Handbook of Statistics-II. North Holland

  • Cano JM, Dimitriadis YA, Arauzo MJ, Coronado J (1996) Fasart: a new neuro-fuzzy architecture for incremental learning in systems identification. In: Proceedings of 13th world congress of IFAC, volume F, pp 133–138

  • Chiu SL (1994) Fuzzy Model Identification Based on Clustering estimation. Intell Fuzzy Syst 2:267–278

    Article  MathSciNet  Google Scholar 

  • Crone SF, Kourentzes N (2010) Feature selection for time series prediction - a combined filter and wrapper approach for neural networks. Neurocomputing 73(10–12):1923–1936

    Article  Google Scholar 

  • Ding C, Peng H (2005) Minimum redundancy feature selection from microarray gene expression data. J Bioinform Comput Biol 3(2):185–205

    Article  MathSciNet  Google Scholar 

  • Elman JL (1990) Finding Structure in Time. Cognitive Sci 10:179–211

    Article  Google Scholar 

  • Estévez PA., Tesmer M, Perez CA, Zurada JM (2009) Normalized mutual information feature selection. IEEE T Neural Networ 20(2):189–201

    Article  Google Scholar 

  • Fakhraei S, Soltanian-Zadeh H, Fotouhi F, Elisevich K (2010a) Consensus feature ranking in datasets with missing values. In: Proceedings—9th International Conference on Machine Learning and Applications, ICMLA 2010, p 771–775

  • Fakhraei S, Soltanian-Zadeh H, Fotouhi F, Elisevich K (2010b) Effect of classifiers in consensus feature ranking for biomedical datasets. In: International Conference on Information and Knowledge Management, Proceedings, pp 67–68

  • Garcia S, Fernandez A, Luengo J, Herrera F (2010) Advanced nonparametric tests for multiple comparisons in the design of experiments in computational intelligence and data mining: Experimental analysis of power. Inf Sci 180(10):2044–2064

    Article  Google Scholar 

  • García-Pajares R, Benítez JM, Palmero GS (2008) Feature selection for time series forecasting: a Case Study. Eighth International conference on hybrid intelligent systems, p 555–560, 2008

  • Guyon I, Elisseeff A (2003) An Introduction to Variable and Feature Selection. J Mach Learn Res 3:1157–1182

    MATH  Google Scholar 

  • Hall MA (2000) Correlation-based feature selection for discrete and numeric class machine learning. In: Proc. 17th International Conf. Machine Learning, pp 359–366

  • Hashemi HB, Yazdani N, Shakery A, Naeini MP (2010) Application of ensemble models in web ranking. In: 5th International Symposium on Telecommunications, IST 2010, pp 726–731

  • He W, Wang Z, Jiang H (2008) Model optimizing and feature selecting for support vector regression in time series forecasting. Neurocomputing 72(1–3):600–611

    Article  Google Scholar 

  • Ji Y, Hao J, Reyhani N, Lendasse A (2005) Direct and recursive prediction of time series using mutual information. Lect Notes Comput Sc 3512:1010–1017

    Article  Google Scholar 

  • Khazaee PR, Mozayani N, Motlagh MRJ (2008) A genetic-based input variable selection algorithm using mutual information and wavelet network for time series prediction. In: Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, number 4811607, pp 2133–2137

  • Kira K, Rendell LA (1992) The feature selection problem: traditional methods and new algorithm. In: Proceedings AAAI-92, pp 129–134

  • Liu H, Motoda H (1998) Feature Selection for Knowledge Discovery and Data Mining. Springer

  • Liu H, Motoda H (2008) Computational Methods of Feature Selection. Chapman & Hall/CRC

  • Loftsgaarden D, Queensberry G (1995) A nonparametric estimate of a multivariate density function. Ann Math Stat 36:1049–1051

    Google Scholar 

  • MacQueen JB (1967) Some methods for classification and analysis of multivariate observations. In: Proceedings of 5-th Berkeley Symposium on Mathematical Statistics and Probability, pp 281–297

  • Maldonado S, Weber R, Basak J (2001) Simultaneous feature selection and classification using kernel-penalized support vector machines. Inf Sci 181(1):115–128

    Article  Google Scholar 

  • Mitchell TM (1997) Machine Learning. McGraw-Hill, New York

    MATH  Google Scholar 

  • Nelles O (2001) Nonlinear system identification. Springer

  • Neumayer R, Nørvåg K (2011) Evaluation of feature combination approaches for text categorisation. Lect Notes Comput Sc 6804:438–448

    Article  Google Scholar 

  • Peng H, Long F, Ding C (2005) Feature selection based on mutual information: criteria of max-dependency, max-relevance, and min-redundancy. IEEE Trans Pattern Anal Mach Intell 27(8):1226–1238

    Article  Google Scholar 

  • Quinlan JR (1986) Induction of decision trees. Mach Learn 1(1):81–106

    Google Scholar 

  • Quinlan JR (1993) C4.5: programs for machine learning. Morgan Kaufmann Publishers Inc., San Fransisco

  • Robnik-Sikonja M, Kononenko I (1997) An adaptation of Relief for attribute estimation in regression. In: Proceedings of the Fourteenth International Conference on Machine Learning ICML, pp 296–304

  • Rubio G, Herrera LJ, Pomares H, Rojas I, Guillén A (2010) Design of specific-to-problem kernels and use of kernel weighted K-nearest neighbours for time series modelling. Neurocomputing 73(10–12):1965–1975

    Article  Google Scholar 

  • Saeys Y, Abeel T, Van De Peer Y (2008) Robust feature selection using ensemble feature selection techniques. Lect Notes Comput Sc (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 5212 LNAI(PART 2):313–325

    Google Scholar 

  • Sakai T (2003) Average gain ratio: a simple retrieval performance measure for evaluation with multiple relevance levels. In Proocedings of the 26th annual international ACM SIGIR conference on Research and development in information retrieval, pp 417–418

  • Shannon C, Weaver W (1948) The mathematical theory of communication. University of Illinois Press

  • Shen K, Ong C, Li X, Hui Z, Wilder-Smith EPV (2007) A feature selection method for multilevel mental fatigue EEG classification. IEEE T Biomed Eng 54:1231–1237

    Article  Google Scholar 

  • Sorjamaa A, Hao J, Reyhani N, Ji Y, Lendasse A (2007) Methodology for long-term prediction of time series. Neurocomputing 70:2861–2869

    Article  Google Scholar 

  • Tripoliti EE, Fotiadis DI, Argyropoulou M (2001) A supervised method to assist the diagnosis and monitor progression of Alzheimer’s disease using data from an fMRI experiment. Artif Intell Med 53(1):35–45

    Article  Google Scholar 

  • Tripoliti E.E., Fotiadis D.I., Argyropoulou M., Manis G. (2010) A six stage approach for the diagnosis of the Alzheimer’s disease based on fMRI data. J Biomed Inform 43(2):307–320

    Article  Google Scholar 

  • Uncu O, Turksen IB (2007) A novel feature selection approach: combining feature wrappers and filters. Inf Sci 177(2):449–466

    Article  MathSciNet  MATH  Google Scholar 

  • Vapnik VN (1995) The nature of statistical learning theory. Springer Verlag

  • Voukantsis D, Niska H, Karatzas K, Riga M, Damialis A, Vokou D (2010) Forecasting daily pollen concentrations using data-driven modeling methods in Thessaloniki, Greece. Atmos Environ 44(39):5101–5111

    Article  Google Scholar 

  • Wang H, Khoshgoftaar TM, Napolitano A (2010) A comparative study of ensemble feature selection techniques for software defect prediction. In: Proceedings—9th International Conference on Machine Learning and Applications, ICMLA 2010, pp 135–140

  • Wilcoxon F (1945) Individual comparisons by ranking methods. Biometrics 1:80–83

    Article  Google Scholar 

  • Witten IH, Frank E (2011) Data mining: practical machine learning tools and techniques, 3rd edn. Morgan Kaufmann, San Fransisco

  • Yan W (2007) Fusion in multi-criterion feature ranking. In: Fusion 2007 10th International Conference on Information Fusion

  • Yan W, Qiu H, Xue Y (2009) Gaussian process for long-term time-series forecasting. In: Proceedings of the international joint conference on neural networks, number 5178729, pp 3420–3427

  • Yu L, Liu H (2004) Efficient feature selection via analysis of relevance and redundancy. J Mach Learn Res, 5(1):1205–1244

    MATH  Google Scholar 

Download references

Acknowledgments

This work was partially supported by the Spanish Ministry of Science and Innovation (MICINN) under grants no. TIN-2009-14575 and DPI2009-14410-C02-02. The authors acknowledge the effectiveness of the comments by anonymous referees, which have helped to improve the quality of the paper.

Author information

Authors and Affiliations

  1. Department of Energy and Computer and Information Technologies, Fundación CARTIF, Boecillo, Spain

    Rubén García-Pajares & G. I. Sainz-Palmero

  2. Department Computer Science and Artificial Intelligence, CITIC-UGR, Universidad de Granada, Granada, Spain

    José M. Benítez

  3. Department of Systems Engineering and Control, School of Industrial Engineering, Universidad de Valladolid,  Valladolid, Spain

    G. I. Sainz-Palmero

Authors
  1. Rubén García-Pajares
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. José M. Benítez
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. G. I. Sainz-Palmero
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Rubén García-Pajares.

Additional information

Communicated by F. Herrera.

Rights and permissions

Reprints and permissions

About this article

Cite this article

García-Pajares, R., Benítez, J.M. & Sainz-Palmero, G.I. FRASel: a consensus of feature ranking methods for time series modelling. Soft Comput 17, 1489–1510 (2013). https://doi.org/10.1007/s00500-012-0961-y

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00500-012-0961-y

Keywords

Search

Navigation