Graph Based Semi-Supervised Learning Method for Imbalanced Dataset

Abstract:

In real application areas, the dataset used may be highly imbalanced and the number of instances for some classes are much higher than that of the other classes. When learning from highly imbalanced dataset, the classifier tends to be adapted to suit the majority class, which might make classifier to obtain a high predictive accuracy over the majority class, but poor accuracy over the minority class. To solve this problem, we put forward a novel graph based semi-supervised learning method for imbalanced dataset, called GSMID. GSMID characterize the class equilibrium constraint as the smoothness of class labels. It’s expected to derive the optimal assignment of class membership to unlabeled samples by maximizing the correlations of classes and simultaneously as smooth as possible on instance graph. The experiments comparing GSMID to SVM and other graph based semi-supervised learning methods on several real-world datasets show GSMM can effectively improve the classification accuracy on imbalanced dataset, especially when data is highly skewed.

• Periodical: Applied Mechanics and Materials (Volumes 556-562)
• Pages: 4040-4044
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.556-562.4040
• Online since: May 2014
• Authors: Chen Guang Zhang, Yan Zhang*, Xia Huan Zhang
• Keywords: Class Correlation, Graph Semi-Supervised Learning, Imbalanced Dataset

References

[1] V. López, etc.: An insight into classification with imbalanced data: Empirical results and current trends on using data intrinsic characteristics, Information Sciences. Vol. 250(20) (2013), p.113–141.

DOI: 10.1016/j.ins.2013.07.007

[2] A. Fernández, etc.: Hierarchical fuzzy rule based classification systems with genetic rule selection for imbalanced data-sets, Inter. J. of App. Reasoning. Vol. 50(3)(2009), pp.561-577.

DOI: 10.1016/j.ijar.2008.11.004

[3] R. Batuwita, V. Palade, FSVM-CIL: fuzzy support vector machines for class imbalance learning, IEEE Transactions on Fuzzy Systems. Vol. 18 (3) (2010), pp.558-571.

DOI: 10.1109/tfuzz.2010.2042721

[4] F. Herrera: Genetic fuzzy systems: taxonomy, current research trends and prospects, Evolutionary Intelligence, Vol. 1(1)(2008), pp.27-46.

DOI: 10.1007/s12065-007-0001-5

[5] R. Batuwita, V. Palade: Proc. of the 2010 International Joint Conference on Neural Networks (Barcelona, Spain, July 18-23, 2010). pp.1-8.

[6] M. Galar and etc.: A review on ensembles for class imbalance problem: bagging, boosting and hybrid based approaches, IEEE Trans. on Sys., Man, and Cyber. Vol. 42 (4) (2012), pp.463-484.

DOI: 10.1109/tsmcc.2011.2161285

[7] Y. LIN, Y. LEE,G. WAHBA: Support Vector Machines for Classification in Nonstandard Situations, March. Learn. Vol. 46(2002), pp.191-202.

[8] M. Alberto, B. Matteo, R.G. Valentini: A neural network algorithm for semi-supervised node label learning from unbalanced data, Neural Networks. Vol. 43(2013), p.84–98.

DOI: 10.1016/j.neunet.2013.01.021

[9] J. Wang, T. Jebara, S.F. Chang: Semi-supervised learning using greedy max-cut, The Journal of Machine Learning Research. Vol. 14(1)(2013), pp.771-800.

[10] T.M. Huang, V. Kecman: Knowledge Based and Emergent Technologies Relied Intelligent Information and Engineering Systems(Springer Verlag, Heidelberg 2004).

[11] C.G. Zhang, Y.J. Li: Hash graph based semi-supervised learning method and its application in image segmentation, Acta Automatica Sinica. Vol. 36 (11)(2010), pp.1527-1533.

DOI: 10.3724/sp.j.1004.2010.01527

[12] C.C. Chang, C.J. Lin: LIBSVM: a library for support vector machines. http: / www. csie. ntu. edu. tw/~cjlin/libsvm.

[13] Asuncion A, Newman D. UCI machine learning repository. http: /archive. ics. uci. edu/ml.