Cost-sensitive sample shifting in feature space

Abstract

The asymmetry of different misclassification costs is a common problem in many realistic applications. As one of the most familiar preprocessing methods, cost-sensitive resampling has drawn great attention due to its easy-implemented and universal properties. However, current methods mainly concentrate on changing the amount of the training set, which will alter the original distribution shapes and lead to the classifiers be over-fitted or unstable. For this case, a new method named cost-sensitive kernel shifting is proposed. The training data are remapped from the input space to the feature space by a particular kernel function, in which a distance metric is defined. Then the outliers are eliminated and the informative samples, including border and edge samples are selected due to the neighbor and geometrical information in the mapped space. Thirdly the positions of all the selected samples in the feature space are shifted. A moving step length is defined in proportion to both the ratio and different of the misclassification costs. In all steps only the kernel matrix is needed to be reshaped due to the kernel trick. Experiments on both synthetic and public datasets verify the effectiveness of the proposed methods.

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

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Notes

  1. 1.

    In fact, no shitting is needed since all the data are exactly classified.

  2. 2.

    http://prtools.tudelft.nl/.

References

  1. 1.

    Khan SH, Hayat M, Bennamoun M et al (2018) Cost-sensitive learning of deep feature representations from imbalanced data. IEEE Trans Neural Netw Learn Syst 29(8):3573–3587

    Article  Google Scholar 

  2. 2.

    García V, Sánchez JS, Marqués AI et al (2019) Understanding the apparent superiority of over-sampling through an analysis of local information for class-imbalanced data. Expert Syst Appl. https://doi.org/10.1016/j.eswa.2019.113026

    Article  Google Scholar 

  3. 3.

    Kulluk S, Ozbakir L, Tapkan PZ et al (2016) Cost-sensitive meta-learning classifiers: MEPAR-miner and DIFACONN-miner. Knowl-Based Syst 98:148–161

    Article  Google Scholar 

  4. 4.

    Zhang G, Sun H, Ji Z et al (2016) Cost-sensitive dictionary learning for face recognition. Pattern Recogn 60:613–629

    Article  Google Scholar 

  5. 5.

    Turney PD (2000) Types of cost in inductive concept learning. In: Proceedings of the ICML’2000 workshop on cost-sensitive learning. Stanford University, IEEE Press, pp. 15–21

  6. 6.

    Zhou ZH, Liu XY (2006) On multi-class cost-sensitive learning. In: Proceedings, the twenty-first national conference on artificial intelligence and the eighteenth innovative applications of artificial intelligence conference, July 16–20, 2006, Boston, Massachusetts, USA. AAAI Press

  7. 7.

    Zhao Z, Wang X (2018) Cost-sensitive SVDD models based on a sample selection approach. Appl Intell 48:4247–4266

    Article  Google Scholar 

  8. 8.

    Sheng VS, Ling CX (2006) Thresholding for making classifiers cost-sensitive. In: Proceedings of the 21st national conference on artificial intelligence. Boston, Massachusetts, pp 476–481

  9. 9.

    Dhar S, Cherkassky V (2015) Development and evaluation of cost-sensitive universum-SVM. IEEE Trans Cybern 45(4):806–818

    Article  Google Scholar 

  10. 10.

    Datta S, Das S (2015) Near-Bayesian support vector machines for imbalanced data classification with equal or unequal misclassification costs. Neural Netw 70:39–52

    Article  Google Scholar 

  11. 11.

    Zhu H, Wang X (2017) A cost-sensitive semi-supervised learning model based on uncertainty. Neurocomputing 251:106–114

    Article  Google Scholar 

  12. 12.

    Zhang L, Zhang D (2017) Evolutionary cost-sensitive extreme learning machine. IEEE Trans Neural Netw Learn Syst 28(12):3045–3060

    MathSciNet  Article  Google Scholar 

  13. 13.

    Cheng F, Zhang J, Wen C (2016) Cost-sensitive large margin distribution machine for classification of imbalanced data. Pattern Recogn Lett 80:107–112

    Article  Google Scholar 

  14. 14.

    Turney PD (1995) Cost-sensitive classification: empirical evaluation of a hybrid genetic decision tree induction algorithm. J Artif Intell Res 2:369–409

    Article  Google Scholar 

  15. 15.

    Li X, Zhao H, Zhu W (2015) A cost sensitive decision tree algorithm with two adaptive mechanisms. Knowl-Based Syst 88:24–33

    Article  Google Scholar 

  16. 16.

    Jia X, Liao W et al (2013) Minimum cost attribute reduction in decision-theoretic rough set models. Inf Sci 219:151–167

    MathSciNet  Article  Google Scholar 

  17. 17.

    Shu W, Shen H (2016) Multi-criteria feature selection on cost-sensitive data with missing values. Pattern Recogn 51:268–280

    Article  Google Scholar 

  18. 18.

    Tapkana P, Özbakıra L, Kulluka S et al (2016) A cost-sensitive classification algorithm: BEE-Miner. Knowl-Based Syst 95:99–113

    Article  Google Scholar 

  19. 19.

    Domingos P (1999) Metacost: a general method for making classifiers cost-sensitive. In: Proceedings of the fifth ACM SIGKDD international conference on knowledge discovery and data mining. San Diego, CA, ACM Press, pp 155–164

  20. 20.

    Kima YJ, Baik B, Cho S (2016) Detecting financial misstatements with fraud intention using multi-class cost-sensitive learning. Expert Syst Appl 62:32–43

    Article  Google Scholar 

  21. 21.

    Zhao Z, Wang X (2018) Multi-segments Naïve Bayes classifier in likelihood space. IET Comput Vision 12(6):882–891

    Article  Google Scholar 

  22. 22.

    Elkan C (2001) The foundations of cost-sensitive learning. In: Proceedings of 17th international joint conference on artificial intelligence, pp. 973–978

  23. 23.

    Zadrozny B, Langford J, Abe N (2003) Cost-sensitive learning by cost-proportionate example weighting. In: Proceedings of the third IEEE international conference on data mining (ICDM’03)

  24. 24.

    K.M. Ting. Inducing cost-sensitive trees via instance weighting. In The 2nd European Symposium on Principles of KDD, Springer-Verlag, 1998, 139–147

  25. 25.

    Ting KM (2002) An instance-weighting method to induce cost-sensitive trees. IEEE Trans Knowl Data Eng 14(3):659–665

    MathSciNet  Article  Google Scholar 

  26. 26.

    Zhang XL (2017) Speech separation by cost-sensitive deep learning. In: Proceedings of APSIPA annual summit and conference, 12–15 December, 2017, Malaysia

  27. 27.

    Choi S, Kim YJ, Briceno S, et al. (2017) Cost-sensitive prediction of airline delays using machine learning. In: 36th IEEE/AIAA digital avionics systems conference (DASC), SEP 17-21

  28. 28.

    Zhang J, Cui X, Li J et al (2017) Imbalanced classification of mental workload using a cost-sensitive majority weighted minority oversampling strategy. Cogn Tech Work 19:633–653

    Article  Google Scholar 

  29. 29.

    Sun Z, Song Q, Zhu X et al (2015) A novel ensemble method for classifying imbalanced data. Pattern Recogn 48:1623–1637

    Article  Google Scholar 

  30. 30.

    Wang S, Li Z, Chao W, et al. (2012) Applying adaptive over-sampling technique based on data density and cost-sensitive SVM to imbalanced learning. In: IEEE world congress on computational intelligence, June, 10–15, 2012, Brisbane, Australia

  31. 31.

    Han H, Wang WY, Mao BH (2005) Borderline-SMOTE: a new over-sampling method in imbalanced data sets learning. Advances in intelligent computing. Springer, Berlin Heidelberg

    Google Scholar 

  32. 32.

    Ertekin S, Huang J, Bottou L, et al. (2007) Learning on the border: active learning in imbalanced data classification. In: Proceedings of the sixteenth ACM conference on information and knowledge management, CIKM 2007, Lisbon, Portugal, November 6–10, 2007 ACM

  33. 33.

    Borowska K, Stepaniuk J (2016) Imbalanced data classification: a novel re-sampling approach combining versatile improved SMOTE and rough sets. Computer Information Systems and Industrial Management. In: CISIM 2016. Lecture Notes in Computer Science, vol 9842. Springer, Cham

  34. 34.

    Wang Q, Luo Z, Huang J, et al. (2017) A novel ensemble method for imbalanced data learning: bagging of extrapolation-SMOTE SVM. In: Computational Intelligence and Neuroscience, 2017, Article ID 1827016

  35. 35.

    Nanni L, Fantozzi C, Lazzarini N (2015) Coupling different methods for overcoming the class imbalance problem. Neurocomputing 158:48–61

    Article  Google Scholar 

  36. 36.

    Perez-Ortiz M, Gutierrez PA, Hervas-Martinez C (2013) Borderline Kernel based over-sampling. In: Proceeding International Conference Hybrid Artifical Intelligence System, Salamanca, Spain, September 2013, pp 472–481

  37. 37.

    Mathew J, Luo M, Pang CK, et al (2015) Kernel-based SMOTE for SVM classification of imbalanced datasets. In: Proceedings IEEE IECON, Yokohama, Japan, November 2015, pp 1127–1132

  38. 38.

    Mathew J, Pang CK, Luo M et al (2018) Classification of imbalanced data by oversampling in kernel space of support vector machines. IEEE Trans Neural Netw Learn Syst 29(9):4065–4076

    Article  Google Scholar 

  39. 39.

    Pes B (2020) Learning from high-dimensional biomedical datasets: the issue of class imbalance. IEEE Access 8:13527–13540

    Article  Google Scholar 

  40. 40.

    Kamalov F (2020) Kernel density estimation based sampling for imbalanced class distribution. Inf Sci 512:1192–1201

    MathSciNet  Article  Google Scholar 

  41. 41.

    Liu B, Tsoumakas G (2020) Dealing with class imbalance in classifier chains via random undersampling. Knowl-Based Syst 192:105292

    Article  Google Scholar 

  42. 42.

    Pouyanfar S, Chen S-C, Shyu M-L (2018) Deep spatio-temporal representation learning for multi-class imbalanced data classification. In: IEEE international conference on information reuse and integration for data science, pp 386–393

  43. 43.

    Huang C, Li Y, Loy CC, et al. (2016) Learning deep representation for imbalanced classification. In: IEEE Conference on computer vision and pattern recognition, (CVPR 2016), pp. 5375–5384

  44. 44.

    Wong ML, Seng K, Wong PK (2020) Cost-sensitive ensemble of stacked denoising autoencoders for class imbalance problems in business domain. Expert Syst Appl 141:112918

    Article  Google Scholar 

  45. 45.

    Datta S, Das S (2019) Multiobjective support vector machines: handling class imbalance with pareto optimality. IEEE Trans Neural Netw Learn Syst 30(5):1602–1608

    MathSciNet  Article  Google Scholar 

  46. 46.

    Mullick SS, Datta S, Das S (2019) Generative adversarial minority oversampling. arXiv: 1903.09730

  47. 47.

    Datta S, Nag S, Das S (2019) Boosting with lexicographic programming: addressing class imbalance without cost tuning. IEEE Transactions on Knowledge and Data Engineering. https://doi.org/10.1109/tkde.2019.2894148

    Article  Google Scholar 

  48. 48.

    Mullick SS, Datta S, Dhekane SG et al (2020) Appropriateness of performance indices for imbalanced data classification: an analysis. Pattern Recogn 102:107197

    Article  Google Scholar 

  49. 49.

    Jian C, Gao J, Ao Y (2016) A new sampling method for classifying imbalanced data based on support vector machine ensemble. Neurocomputing 193:115–122

    Article  Google Scholar 

  50. 50.

    Lin WC, Tsai CF, Hu YH et al (2017) Clustering-based undersampling in class-imbalanced data. Inf Sci 409–410:17–26

    Article  Google Scholar 

  51. 51.

    Rayhan F, Ahmed S, Mahbub A, et al (2017) CUSBoost: cluster-based under-sampling with boosting for imbalanced classification. In: 2nd IEEE international conference on computational systems and information technology for sustainable solutions. pp, 70–75

  52. 52.

    Ofek N, Rokach L, Stern R et al (2017) Fast-CBUS: a fast clustering-based undersampling method for addressing the class imbalanced problem. Neurocomputing 243:88–102

    Article  Google Scholar 

  53. 53.

    Nejatian S, Parvin H, Faraji E (2018) Using sub-sampling and ensemble clustering techniques to improve performance of imbalanced classification. Neurocomputing 276:55–66

    Article  Google Scholar 

  54. 54.

    Lu W, Li Z, Chu J (2017) Adaptive ensemble undersampling-Boost: a novel learning framework for imbalanced data. J Syst Softw 132:272–282

    Article  Google Scholar 

  55. 55.

    Sun B, Chen H, Wang J et al (2018) Evolutionary under-sampling based bagging ensemble method for imbalanced data classification. Front Comput Sci. 12:331. https://doi.org/10.1007/s11704-016-5306-z

    Article  Google Scholar 

  56. 56.

    He H, Bai Y, Garcia EA, et al. (2008) ADASYN: adaptive synthetic sampling approach for imbalanced learning. In: Proceeding of international joint conferwence neural network. June 2008, pp 1322–1328

  57. 57.

    Jiang L, Qiu C, Li C (2015) A novel minority cloning technique for cost-sensitive learning. Int J Pattern Recognit Artif Intell 29(4):1551004

    Article  Google Scholar 

  58. 58.

    Ahmed S, Mahbub A, Rayhan F, et al. (2017) Hybrid methods for class imbalance learning employing Bagging with sampling techniques. In: 2017 2nd international conference on computational systems and information technology for sustainable solution (CSITSS), Bangalore, pp 1–5

  59. 59.

    Qiu C, Jiang L, Kong G (2015) A differential evolution-based method for class-imbalanced cost-sensitive learning. In: 2015 International joint conference on neural networks (IJCNN), Killarney, pp 1–8

  60. 60.

    Zhang C, Song J, Gao W, et al. (2016) An imbalanced data classification algorithm of improved autoencoder neural network. In: 8th international conference on advanced computational intelligence, Chiang Mai, Thailand, February 14–16

  61. 61.

    Wang J, Huang PL, Sun KW, et al. (2013) Ensemble of cost-sensitive hypernetworks for class-imbalanced learning. In: 2013 IEEE international conference on systems, man, and cybernetics

  62. 62.

    Li YH, Maguire L (2011) Selecting critical patterns based on local geometrical and statistical information. IEEE Trans Pattern Anal Mach Intell 33(6):1189–1201

    Article  Google Scholar 

  63. 63.

    Paredes R, Vidal E (2006) Learning prototypes and distances: a prototype reduction technique based on nearest neighbor error minimization. Pattern Recogn 39:180–188

    Article  Google Scholar 

  64. 64.

    Nanni L, Franco A (2011) Reduced Reward-punishment editing for building ensembles of classifiers. Expert Syst Appl 38:2395–2400

    Article  Google Scholar 

  65. 65.

    Alcalá-Fdez J, Fernandez A, Luengo J et al (2011) KEEL data-mining software tool: data set repository, integration of algorithms and experimental analysis framework. J Multiple-Valued Logic Soft Comput 17(2–3):255–287

    Google Scholar 

  66. 66.

    Zhao Z, Wang X, Wu C, Lei L (2019) Hunting optimization: an new framework for single objective optimization problems. IEEE Access 7:31305–31320

    Article  Google Scholar 

Download references

Acknowledgements

This work is supported by the National Natural Science Foundation of China under Grants 61876189, 61273275, 61806219, 61703426 and 61503407.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Xiaodan Wang.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Zhao, Z., Wang, X., Wu, C. et al. Cost-sensitive sample shifting in feature space. Pattern Anal Applic (2020). https://doi.org/10.1007/s10044-020-00890-9

Download citation

Keywords

  • Cost-sensitive learning
  • Resampling
  • Feature space
  • SVM