Developing a precise and consistent model for classifying imbalanced medical data is one of the major challenges in machine learning and data mining. As the advanced growth in medical technology, a classy medical classification system is essential that make use of data mining algorithms to support medical diagnosis practice. Though the standard medical data seldom obeys the requirements of different knowledge engineering tools, most of the medical datasets are considered to be highly imbalanced with respect to their class label. So the imbalancing problem has been found to thwart the efficiency of the learning model. The only way to avoid this problem is to reduce the gap between both majority and minority class instances. In our approach a fuzzy gravitational classifier with weighting scheme is employed, in which weight is optimized using Particle swarm optimization algorithm. The technique is implemented and tested with three well known bench mark imbalanced dataset from UCI and KEEL repository. A comparative study is made with two existing classification methods viz. Weighted nearest neighbour and class based weighted nearest neighbour. Evaluation results shows our hybrid approach gives better performance on imbalanced data in terms of AUC, F-measure and G-mean.

Imbalanced data, PSO optimization, Data gravitation classifier, Fuzzy soft set, JFIM

[1]. R.C. Holte, L. Acker, and B.W. Porter, “Concept Learning and theProblem of Small Disjuncts” Proc. Int’l J. Conf. ArtificialIntelligence, pp. 813-818, 1989.
[2]. G.M. Weiss, “Mining with rarity: a unifying framework”, SIGKDD Explor. Vol.6, pp. 7–19, 2004.
[3]. V. García, R. Mollineda, J.S. Sánchez, “On the k-NN performance in a challenging scenario of imbalance and overlapping”, Pattern Anal. Appl. Vol.11, pp. 269–280, 2008.
[4]. C. Seiffert, T.M. Khoshgoftaar, J. Van Hulse, et al. ,”An empirical study of the classification performance of learners on imbalanced and noisy softwarequality data”, in: Procedings of IEEE International Conference on Information Reuse and Integration, pp. 651–658, 2007.
[5]. K. Napierala, J. Stefanowski, S. Wilk, “Learning from imbalanced data in presence of noisy and borderline examples”, in: Proceedings of 7th International Conference on Rough Sets and Current Trends in Computing (RSCTC2010), pp. 158–167, 2010.
[6]. L. Peng, H. Zhang, B. Yang, Y. Chen, M.T. Qassrawi, G. Lu, “Traffic identification using flexible neural trees”, in: Proceeding of the 18th International Workshop of QoS (IWQoS 2012), pp. 1–5, 2012.
[7]. Fawcett, T., & Provost, F. J. “Adaptive fraud detection. Data Mining and Knowledge Discovery”, Vol.1, no.3, pp.291–316, 1997.
[8]. P. Clifton, A. Damminda, and L. Vincent, “Minority Report in Fraud Detection: Classification of Skewed Data,” ACM SIGKDDExplorations Newsletter, vol. 6, no. 1, pp. 50-59, 2004.
[9]. T. Basu, C.A. Murthy, “Towards enriching the quality of k-nearest neighbour rule for document classification”, Int. J. Mach. Learn. Cybern., pp. 1–9, 2013.
[10]. Ghosh, A., Pal, N., & Das, J.,” A fuzzy rule based approach to cloud cover estimation. Remote Sensing of Environment”, vol.100, no.4, pp. 531–549, 2006.
[11]. Huang, Y. M., Hung, C. M., & Jiau, H. C.,” Evaluation of neural networks and data mining methods on a credit assessment task for class imbalance problem.Nonlinear Analysis: Real World Applications”, vol.7, no.4, pp.720–747, 2006.
[12]. I. Chairi, S. Alaoui, A. Lyhyaoui,, “Intrusion detection based sample selection for imbalanced data distribution”, in: Proceeding of Second International Conference on Innovative Computing Technology (INTECH), pp. 259–264, 2012.
[13]. Xu, L., Chow, M. Y., & Taylor, L. S. (2007), ”Power distribution fault causeidentification with imbalanced data using the data mining-based fuzzyclassification E-algorithm”, IEEE Transactions on Power Systems, vol.22, no.1, pp. 164–171, 2007.
[14]. Chi, Z., Yan, H., & Pham, T. , “Fuzzy algorithms with applications to image processing and pattern recognition”. World Scientific journal off science, pp. 132-140, 1996.
[15]. Mazurowski, M., Habas, P., Zurada, J., Lo, J., Baker, J., & Tourassi, G. (2008),”Training neural network classifiers for medical decision making: The effects of imbalanced datasets on classification performance”, Neural Networks, vol.21, no.2, 427–436, 2008.
[16]. R. Barandela, J.S. Sánchez, V. García, E. Rangel, “Strategies for learning in class imbalance problems”, Pattern Recogn, Vol.36, pp. 849–851, 2003.
[17]. G.E.A.P.A. Batista, R.C. Prati, M.C. Monard, “A study of the behaviour of several methods for balancing machine learning training data”, SIGKDD Explor. Vol.6, pp. 20–29, 2004.
[18]. S. A. Dudani, “The distance-weighted k-nearest-neighbor rule,” IEEE Trans. Syst., Man, Cybern. B, Cybern., vol. SMC-6, no. 4, pp. 325–327, 1976.
[19]. Q. Gao and Z. Wang, “Center-based nearest neighbor classifier,” Pattern Recognit., vol. 40, no. 1, pp. 346–349, 2007.
[20]. J.Wang, P. Neskovic, and L. N. Cooper, “Improving nearest neighbor rule with a simple adaptive distance measure,” Pattern Recognit. Lett., vol. 28, no. 2, pp. 207–213, 2007.
[21]. R. Paredes and E. Vidal, “A class-dependent weighted dissimilarity measure for nearest neighbor classification problems,” Pattern Recognit. Lett., vol. 21, no. 12, pp. 1027–1036, 2000.
[22]. R. Paredes and E. Vidal, “Learning weighted metrics to minimize nearest neighbor classification error,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 28, no. 7, pp. 1100–1110, 2006.
[23]. L. Peng, B. Peng, Y. Chen, and A. Abraham, “Data gravitation based classification,” Inf. Sci., vol. 179, no. 6, pp. 809–819, 2009.
[24]. Sinciya P.O, J. Jeya A Celin, “Weight Optimized Gravitational Classifier for High Dimensional Numerical Data Classification “, International Journal of Pure and Applied Mathematics”, Volume 116, No. 22, pp.251-263, 2017.
[25]. Sinciya P.O, Jeyaa Celin Jesu, “JFIM: A Novel Filter Feature Selection approach Using Joint Feature Interaction Maximization”, International journal of intelligent engineering and systems, Vol.10, No.6, pp. 195-203,2017.
[26]. H. Dubey and V. Pudi. “Class based weighted k nearest neighbor over imbalanced dataset”, PAKDD 2013, pp. 305-316, 2013.