Knowledge and Information Systems

, Volume 32, Issue 1, pp 77–108 | Cite as

On the choice of the best imputation methods for missing values considering three groups of classification methods

  • Julián LuengoEmail author
  • Salvador García
  • Francisco Herrera
Regular Paper


In real-life data, information is frequently lost in data mining, caused by the presence of missing values in attributes. Several schemes have been studied to overcome the drawbacks produced by missing values in data mining tasks; one of the most well known is based on preprocessing, formerly known as imputation. In this work, we focus on a classification task with twenty-three classification methods and fourteen different imputation approaches to missing values treatment that are presented and analyzed. The analysis involves a group-based approach, in which we distinguish between three different categories of classification methods. Each category behaves differently, and the evidence obtained shows that the use of determined missing values imputation methods could improve the accuracy obtained for these methods. In this study, the convenience of using imputation methods for preprocessing data sets with missing values is stated. The analysis suggests that the use of particular imputation methods conditioned to the groups is required.


Approximate models Classification Imputation Rule induction learning Lazy learning Missing values Single imputation 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Acuna E, Rodriguez C (2004) Classification, clustering and data mining applications. Springer, Berlin, pp 639–648CrossRefGoogle Scholar
  2. 2.
    Alcalá-fdez J, Sánchez L, García S, Jesus MJD, Ventura S, Garrell JM, Otero J, Bacardit J, Rivas VM, Fernández JC, Herrera F (2009) Keel: a software tool to assess evolutionary algorithms for data mining problems. Soft Comput 13(3): 307–318CrossRefGoogle Scholar
  3. 3.
    Asuncion A, Newman D (2007) UCI machine learning repository.
  4. 4.
    Atkeson CG, Moore AW, Schaal S (1997) Locally weighted learning. Artif Intell Rev 11: 11–73CrossRefGoogle Scholar
  5. 5.
    Barnard J, Meng X (1999) Applications of multiple imputation in medical studies: From aids to nhanes. Stat Methods Med Res 8(1): 17–36CrossRefGoogle Scholar
  6. 6.
    Batista G, Monard M (2003) An analysis of four missing data treatment methods for supervised learning. Appl Artif Intell 17(5): 519–533CrossRefGoogle Scholar
  7. 7.
    Bezdek J, Kuncheva L (2001) Nearest prototype classifier designs: an experimental study. Int J Intell Syst 16(12): 1445–1473zbMATHCrossRefGoogle Scholar
  8. 8.
    Broomhead D, Lowe D (1988) Multivariable functional interpolation and adaptive networks. Complex Syst 11: 321–355MathSciNetGoogle Scholar
  9. 9.
    Clark P, Niblett T (1989) The cn2 induction algorithm. Mach Learn J 3(4): 261–283Google Scholar
  10. 10.
    Cohen W (1995) Fast effective rule induction. In: Machine learning: proceedings of the twelfth international conference, pp 1–10Google Scholar
  11. 11.
    Cohen W, Singer Y (1999) A simple and fast and and effective rule learner. In: Proceedings of the sixteenth national conference on artificial intelligence, pp 335–342Google Scholar
  12. 12.
    Cover TM, Thomas JA (1991) Elements of information theory, 2nd edn. Wiley, NYzbMATHCrossRefGoogle Scholar
  13. 13.
    Demšar J (2006) Statistical comparisons of classifiers over multiple data sets. J Mach Learn Res 7: 1–30MathSciNetzbMATHGoogle Scholar
  14. 14.
    Ding Y, Simonoff JS (2010) An investigation of missing data methods for classification trees applied to binary response data. J Mach Learn Res 11: 131–170MathSciNetGoogle Scholar
  15. 15.
    Domingos P, Pazzani M (1997) On the optimality of the simple bayesian classifier under zero-one loss. Mach Learn 29: 103–137zbMATHCrossRefGoogle Scholar
  16. 16.
    Ennett CM, Frize M, Walker CR (2001) Influence of missing values on artificial neural network performance. Stud Health Technol Inform 84: 449–453Google Scholar
  17. 17.
    Fan R-E, Chen P-H, Lin C-J (2005) Working set selection using second order information for training support vector machines. J Mach Learn Res 6: 1889–1918MathSciNetzbMATHGoogle Scholar
  18. 18.
    Farhangfar A, Kurgan LA, Pedrycz W (2007) A novel framework for imputation of missing values in databases. IEEE Trans Syst Man Cybern Part A 37(5): 692–709CrossRefGoogle Scholar
  19. 19.
    Farhangfar A, Kurgan L, Dy J (2008) Impact of imputation of missing values on classification error for discrete data. Pattern Recognit 41(12): 3692–3705zbMATHCrossRefGoogle Scholar
  20. 20.
    Fayyad U, Irani K (1993) Multi-interval discretization of continuous-valued attributes for classification learning. In: Proceedings of 13th international joint conference on uncertainly in artificial intelligence (IJCAI93), pp. 1022–1029Google Scholar
  21. 21.
    Feng H, Guoshun C, Cheng Y, Yang B, Chen Y (2005) A svm regression based approach to filling in missing values. In: Khosla R, Howlett RJ, Jain LC (eds) ‘KES (3)’, vol 3683 of lecture notes in computer science. Springer, Berlin, pp 581–587Google Scholar
  22. 22.
    Frank E, Witten I (1998) Generating accurate rule sets without global optimization. In: Proceedings of the fifteenth international conference on machine learning, pp 144–151Google Scholar
  23. 23.
    García-Laencina P, Sancho-Gómez J, Figueiras-Vidal A (2009) Pattern classification with missing data: a review. Neural Comput Appl. 9(1): 1–12Google Scholar
  24. 24.
    García S, Herrera F (2008) An extension on “statistical comparisons of classifiers over multiple data sets” for all pairwise comparisons. J Mach Learn Res 9: 2677–2694zbMATHGoogle Scholar
  25. 25.
    Gheyas IA, Smith LS (2010) A neural network-based framework for the reconstruction of incomplete data sets. Neurocomputing In Press, Corrected ProofGoogle Scholar
  26. 26.
    Grzymala-Busse J, Goodwin L, Grzymala-Busse W, Zheng X (2005) Handling missing attribute values in preterm birth data sets. In: Proceedings of 10th international conference of rough sets and fuzzy sets and data mining and granular computing(RSFDGrC), pp 342–351Google Scholar
  27. 27.
    Grzymala-Busse JW, Hu M (2000) A comparison of several approaches to missing attribute values in data mining. In: Ziarko W, Yao YY (eds) Rough sets and current trends in computing, vol 2005 of lecture notes in computer science, Springer, pp 378–385Google Scholar
  28. 28.
    Hruschka ER Jr., Hruschka ER, Ebecken NF (2007) Bayesian networks for imputation in classification problems. J Intell Inf Syst 29(3): 231–252CrossRefGoogle Scholar
  29. 29.
    Kim H, Golub GH, Park H (2005) Missing value estimation for dna microarray gene expression data: local least squares imputation. Bioinformatics 21(2): 187–198CrossRefGoogle Scholar
  30. 30.
    Kwak N, Choi C-H (2002) Input feature selection by mutual information based on parzen window. IEEE Trans Pattern Anal Mach Intell 24(12): 1667–1671CrossRefGoogle Scholar
  31. 31.
    Kwak N, Choi C-H (2002) Input feature selection for classification problems. IEEE Trans Neural Netw 13(1): 143–159CrossRefGoogle Scholar
  32. 32.
    Cessie S le, van Houwelingen J (1992) Ridge estimators in logistic regression. Appl Stat 41(1): 191–201zbMATHCrossRefGoogle Scholar
  33. 33.
    Li D, Deogun J, Spaulding W, Shuart B (2004) Towards missing data imputation: a study of fuzzy k-means clustering method. In: Proceedings of 4th international conference of rough sets and current trends in computing (RSCTC), pp 573–579Google Scholar
  34. 34.
    Little RJA, Rubin DB (1987) Statistical analysis with missing data, wiley series in probability and statistics, 1st edn. Wiley, New YorkGoogle Scholar
  35. 35.
    Luengo J, García S, Herrera F (2010) A study on the use of imputation methods for experimentation with Radial Basis Function Network classifiers handling missing attribute values: the good synergy between RBFNs and EventCovering method. Neural Netw 23(3): 406–418CrossRefGoogle Scholar
  36. 36.
    Matsubara ET, Prati RC, Batista GEAPA, Monard MC (2008) Missing value imputation using a semi-supervised rank aggregation approach. In: Zaverucha G, da Costa ACPL (eds) ‘SBIA’, vol 5249 of lecture notes in computer science. Springer, Berlin, pp 217–226Google Scholar
  37. 37.
    McLachlan G (2004) Discriminant analysis and statistical pattern recognition. Wiley, NYzbMATHGoogle Scholar
  38. 38.
    Merlin P, Sorjamaa A, Maillet B, Lendasse A (2010) X-SOM and L-SOM: a double classification approach for missing value imputation. Neurocomputing 73(7–9): 1103–1108CrossRefGoogle Scholar
  39. 39.
    Michalksi R, Mozetic I, Lavrac N (1986) The multipurpose incremental learning system aq15 and its testing application to three medical domains. In: Proceedings of 5th international conference on artificial intelligence (AAAI), pp 1041–1045Google Scholar
  40. 40.
    Moller F (1990) A scaled conjugate gradient algorithm for fast supervised learning. Neural Netw 6: 525–533CrossRefGoogle Scholar
  41. 41.
    Nogueira BM, Santos TRA, Zárate LE (2007) Comparison of classifiers efficiency on missing values recovering: application in a marketing database with massive missing data. In: ‘CIDM’, IEEE, pp 66–72Google Scholar
  42. 42.
    Oba S, aki Sato M, Takemasa I, Monden M, ichi Matsubara K, Ishii S (2003) A bayesian missing value estimation method for gene expression profile data. Bioinformatics 19(16): 2088–2096CrossRefGoogle Scholar
  43. 43.
    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): 1226–1238CrossRefGoogle Scholar
  44. 44.
    Pham DT, Afify AA (2005) Rules-6: a simple rule induction algorithm for supporting decision making. In: Industrial electronics society, 2005. IECON 2005. 31st annual conference of IEEE, pp 2184–2189Google Scholar
  45. 45.
    Pham DT, Afify AA (2006) Sri: A scalable rule induction algorithm. Proc Inst Mech Eng Part C J Mech Eng Sci 220:537–552Google Scholar
  46. 46.
    Plat J (1991) A resource allocating network for function interpolation. Neural Comput 3(2): 213–225CrossRefGoogle Scholar
  47. 47.
    Platt JC (1999) Fast training of support vector machines using sequential minimal optimization. In: Advances in kernel methods: support vector learning. MIT Press, Cambridge, pp 185–208Google Scholar
  48. 48.
    Pyle D (1999) Data preparation for data mining. Morgan Kaufmann, Los AltosGoogle Scholar
  49. 49.
    Qin B, Xia Y, Prabhakar S (2010) Rule induction for uncertain data. Knowl Inf Syst, doi: 10.1007/s10115-010-0335-7, pp 1–28 (in press)
  50. 50.
    Quinlan J (1993) C4.5:programs for machine learning. Morgan Kauffman, Los AltosGoogle Scholar
  51. 51.
    Reddy C, Park J-H (2010) Multi-resolution boosting for classification and regression problems. Knowl Inf Syst, doi: 10.1007/s10115-010-0358-0, pp 1–22, (in press)
  52. 52.
    Saar-Tsechansky M, Provost F (2007) Handling missing values when applying classification models. J Learn Res 8: 1623–1657zbMATHGoogle Scholar
  53. 53.
    Safarinejadian B, Menhaj M, Karrari M (2010) A distributed EM algorithm to estimate the parameters of a finite mixture of components. Knowl Inf Syst 23(3): 267–292CrossRefGoogle Scholar
  54. 54.
    Schafer JL (1997) Analysis of incomplete multivariate data. Chapman & Hall, LondonzbMATHCrossRefGoogle Scholar
  55. 55.
    Schneider T (2001) Analysis of incomplete climate data: estimation of mean values and covariance matrices and imputation of missing values. J Clim 14: 853–871CrossRefGoogle Scholar
  56. 56.
    Song Q, Shepperd M, Chen X, Liu J (2008) Can k-NN imputation improve the performance of C4.5 with small software project data sets? A comparative evaluation. J Syst Softw 81(12): 2361–2370CrossRefGoogle Scholar
  57. 57.
    Troyanskaya O, Cantor M, Sherlock G, Brown P, Hastie T, Tibshirani R, Botstein D, Altman RB (2001) Missing value estimation methods for dna microarrays. Bioinformatics 17(6): 520–525CrossRefGoogle Scholar
  58. 58.
    Twala B (2009) An empirical comparison of techniques for handling incomplete data using decision trees. Appl Artif Intell 23: 373–405CrossRefGoogle Scholar
  59. 59.
    Unnebrink K, Windeler J (n.d.)Google Scholar
  60. 60.
    Wang H, Wang S (2010) Mining incomplete survey data through classification. Knowl Inf Syst 24(2): 221–233CrossRefGoogle Scholar
  61. 61.
    Wilson D (1972) Asymptotic properties of nearest neighbor rules using edited data. IEEE Trans Syst Man Cybern 2(3): 408–421zbMATHCrossRefGoogle Scholar
  62. 62.
    Wong AKC, Chiu DKY (1987) Synthesizing statistical knowledge from incomplete mixed-mode data. IEEE Trans Pattern Anal Mach Intell 9(6): 796–805CrossRefGoogle Scholar
  63. 63.
    Wu X, Urpani D (1999) Induction by attribute elimination. IEEE Trans Knowl Data Eng 11(5): 805–812CrossRefGoogle Scholar
  64. 64.
    Zheng Z, Webb GI (2000) Lazy learning of bayesian rules. Mach Learn 41(1): 53–84CrossRefGoogle Scholar

Copyright information

© Springer-Verlag London Limited 2011

Authors and Affiliations

  • Julián Luengo
    • 1
    Email author
  • Salvador García
    • 2
  • Francisco Herrera
    • 1
  1. 1.Department of Computer Science and Artificial IntelligenceCITIC-University of GranadaGranadaSpain
  2. 2.Dept. of Computer ScienceUniversity of JaénJaénSpain

Personalised recommendations