Skip to main content
SpringerLink
Log in

Improving optimum-path forest learning using bag-of-classifiers and confidence measures

  • Theoretical Advances
  • Published:
Pattern Analysis and Applications Aims and scope Submit manuscript

Abstract

Machine learning techniques have been actively pursued in the last years, mainly due to the great number of applications that make use of some sort of intelligent mechanism for decision-making processes. In this work, we presented an ensemble of optimum-path forest (OPF) classifiers, which consists into combining different instances that compute a score-based confidence level for each training sample in order to turn the classification process “smarter”, i.e., more reliable. Such confidence level encodes the level of effectiveness of each training sample, and it can be used to avoid ties during the OPF competition process. Experimental results over fifteen benchmarking datasets have shown the effectiveness and efficiency of the proposed approach for classification problems, with more accurate results in more than 67% of the datasets considered in this work. Additionally, we also considered a bagging strategy for comparison purposes, and we showed the proposed approach can lead to considerably better results.

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.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

Notes

  1. http://archive.ics.uci.edu/ml.

  2. http://lrs.icg.tugraz.at/research/aflw.

  3. Notice the percentages have been empirically chosen, being more intuitive to provide a larger validating set for calculating the confidence levels.

References

  1. Al-Ani A, Deriche M (2002) A new technique for combining multiple classifiers using the dempster–shafer theory of evidence. J Artif Intell Res 17(1):333–361

    Article  MathSciNet  MATH  Google Scholar 

  2. Allène C, Audibert JY, Couprie M, Keriven R (2010) Some links between extremum spanning forests, watersheds and min-cuts. Image Vis Comput 28(10):1460–1471

    Article  Google Scholar 

  3. Amancio DR, Comin CH, Casanova D, Travieso G, Bruno OM, Rodrigues FA, Costa LF (2014) A systematic comparison of supervised classifiers. PLoS ONE 9(4):e94,137

    Article  Google Scholar 

  4. Amorim WP, Falcão AX, Papa JP, Carvalho MH (2016) Improving semi-supervised learning through optimum connectivity. Pattern Recogn 60:72–85

    Article  Google Scholar 

  5. Andrews DF (1972) Plots of high-dimensional data. Biometrics 28(1):125–136

    Article  Google Scholar 

  6. Breiman L (1996) Bagging predictors. Mach Learn 24(2):123–140

    MathSciNet  MATH  Google Scholar 

  7. Castillo E, Peteiro-Barral D, Berdiñas BG, Fontenla-Romero O (2015) Distributed one-class support vector machine. Int J Neural Syst 25(07):1550,029

    Article  Google Scholar 

  8. Dash JK, Mukhopadhyay S (2016) Similarity learning for texture image retrieval using multiple classifier system. Multimed Tools Appl 1–25. doi:10.1007/s11042-016-4228-y

  9. Demsar J (2006) Statistical comparisons of classifiers over multiple data sets. J Mach Learn Res 7:1–30

    MathSciNet  MATH  Google Scholar 

  10. Fernandes SEN, Scheirer W, Cox DD (2015) Papa JP progress in pattern recognition, image analysis, computer vision, and applications: 20th Iberoamerican congress, CIARP 2015, Montevideo, Uruguay, November 9–12, 2015, Proceedings, chap. improving optimum-path forest classification using confidence measures, pp 619–625. Springer International Publishing, Cham

  11. Fernandes SEN, Souza AN, Gastaldello DS, Pereira DR, Papa JP (2017) Pruning optimum-path forest ensembles using metaheuristic optimization for land-cover classification. Int J Remote Sens 38:5736–5762

    Article  Google Scholar 

  12. Folino G, Pisani FS (2015) Combining ensemble of classifiers by using genetic programming for cyber security applications. Springer International Publishing, Cham, pp 54–66

    Google Scholar 

  13. Giacinto G, Roli F, Fumera G (2000) Selection of classifiers based on multiple classifier behaviour. Springer, Berlin, pp 87–93

    MATH  Google Scholar 

  14. Ho TK (1998) The random subspace method for constructing decision forests. IEEE Trans Pattern Anal Mach Intell 20(8):832–844

    Article  Google Scholar 

  15. Hunter JD (2007) Matplotlib: a 2d graphics environment. Comput Sci Eng 9(3):90–95

    Article  Google Scholar 

  16. Kittler J, Hatef M, Duin RPW, Matas J (1998) On combining classifiers. IEEE Trans Pattern Anal Mach Intell 20(3):226–239

    Article  Google Scholar 

  17. Koziol J, Hacke W (1991) A bivariate version of andrews plots. IEEE Trans Biomed Eng 38(12):1271–1274

    Article  Google Scholar 

  18. Kuncheva L, Skurichina M, Duin RPW (2002) An experimental study on diversity for bagging and boosting with linear classifiers. Inf Fus 3(4):245–258

    Article  Google Scholar 

  19. Kuncheva LI (2004) Combining pattern classifiers: methods and algorithms. Wiley-Interscience, New York

    Book  MATH  Google Scholar 

  20. Nemenyi P (1963) Distribution-free multiple comparisons. Princeton University, Princeton

    Google Scholar 

  21. Papa JP, Falcão AX (2008) A new variant of the optimum-path forest classifier. In: Proceedings of the 4th international symposium on advances in visual computing, Lecture Notes in Computer Science, Springer, Berlin, pp 935–944

  22. Papa JP, Falcão AX (2009) A learning algorithm for the optimum-path forest classifier. In: Torsello A, Escolano F, Brun L (eds) Graph-based representations in pattern recognition, vol 5534. Lecture Notes in Computer Science. Springer, Berlin Heidelberg, pp 195–204

  23. Papa JP, Falcão AX, Albuquerque VHC, Tavares JMRS (2012) Efficient supervised optimum-path forest classification for large datasets. Pattern Recogn 45(1):512–520

    Article  Google Scholar 

  24. Papa JP, Falcão AX, Suzuki CTN (2009) Supervised pattern classification based on optimum-path forest. Int J Imaging Syst Technol 19(2):120–131

    Article  Google Scholar 

  25. Papa JP, Fernandes SEN, Falcão AX (2017) Optimum-path forest based on k-connectivity: theory and applications. Pattern Recogn Lett 87:117–126

    Article  Google Scholar 

  26. Ponti M, Rossi I (2013) Ensembles of optimum-path forest classifiers using input data manipulation and undersampling. Multiple Classif Syst 7872:236–246

    Article  Google Scholar 

  27. Ponti MP, Papa JP (2011) Improving accuracy and speed of optimum-path forest classifier using combination of disjoint training subsets. In: Sansone C, Kittler J, Roli F (eds) Multiple classifier systems, vol 6713. Lecture Notes in Computer Science. Springer, Berlin, pp 237–248

  28. Ponti MP, Papa JP, Levada ALM (2011) A Markov random field model for combining optimum-path forest classifiers using decision graphs and game strategy approach. In: San Martin C, Kim SW (eds) Progress in pattern recognition, image analysis, computer vision, and applications, Lecture Notes in Computer Science, vol 7042, pp 581–590. Springer, Berlin

  29. Souza R, Rittner L, Lotufo RA (2014) A comparison between k-optimum path forest and k-nearest neighbors supervised classifiers. Pattern Recogn Lett 39:2–10

    Article  Google Scholar 

  30. Wilcoxon F (1945) Individual comparisons by ranking methods. Biom Bull 1(6):80–83

    Article  Google Scholar 

  31. Xu L, Krzyzak A, Suen C (1992) Methods of combining multiple classifiers and their applications to handwriting recognition. IEEE Trans Syst Man Cybern 22(3):418–435

    Article  Google Scholar 

  32. Zhang Y, Zhou W, Yuan S (2015) Multifractal analysis and relevance vector machine-based automatic seizure detection in intracranial EEG. Int J Neural Syst 25(6):1550020

    Article  Google Scholar 

Download references

Acknowledgements

The authors are grateful to FAPESP grants #2013/07375-0, #2014/16250-9, #2014/12236-1, and #2016/19403-6, Capes, and CNPq grants #470571/2016-6 and #306166/2014-3 for their financial support. The authors are also grateful to the reviewers for their insightful comments.

Author information

Authors and Affiliations

  1. Department of Computing, Federal University of São Carlos - UFSCar, Rodovia Washington Luís, Km 235 - SP 310, São Carlos, São Paulo, 13565-905, Brazil

    Silas Evandro Nachif Fernandes

  2. Department of Computing, São Paulo State University - UNESP, Av. Eng. Luiz Edmundo Carrijo Coube, 14-01, Bauru, São Paulo, 17033-360, Brazil

    João Paulo Papa

Authors
  1. Silas Evandro Nachif Fernandes
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. João Paulo Papa
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to João Paulo Papa.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Fernandes, S.E.N., Papa, J.P. Improving optimum-path forest learning using bag-of-classifiers and confidence measures. Pattern Anal Applic 22, 703–716 (2019). https://doi.org/10.1007/s10044-017-0677-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10044-017-0677-9

Keywords

Search

Navigation