Abstract
Class imbalance problem comprises of uneven distribution of data/instances in classes which poses a challenge in the performance of classification models. Traditional classification algorithms produce high accuracy rate for majority classes and less accuracy rate for minority classes. Study of such problem is called class imbalance learning. Various methods are used in imbalance learning applications, which modify the distribution of the original dataset by some mechanisms in order to obtain a relatively balanced dataset. Most of the techniques like SMOTE and ADASYN proposed in the literature use oversampling approach to handle class imbalance learning. This paper presents a modified SMOTE approach, i.e., Farthest SMOTE to solve the imbalance problem. FSMOTE approach generates synthetic samples along the line joining the minority samples and its âkâ minority class farthest neighbors. Further, in this paper, FSMOTE approach is evaluated on seven real-world datasets.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
S. Maheshwari, J. Agrawal and S. Sharma, ââNew approach for classification of highly imbalanced datasets using evolutionary algorithms,ââ Int. J. Sci. Eng. Res., vol. 2, no. 7, pp. 1â5, 2011.
A. Amin, S. Anwar, âComparing Oversampling Techniques to Handle the CIP: A Customer Churn Prediction Case Studyâ, IEEE Translations and content mining, Vol. 4, 2016.
G. Weiss, âMining with Rarity: A Unified Frameworkâ, SIGKDD Explorations, Vol. 6, No. 1, pp. 7â19, 2004.
X. Guo, Y. Yin, C. Dong, âOn the class imbalance problemâ, Natural Computation, 2008. ICNCâ08. Fourth International Conference on. 2008.
K. P. N. V. Satyashree, and J. V. R. Murthy, âAn Exhaustive Literature Review on Class Imbalance Problemâ, Int. Journal of Emerging Trends and Technology in Computer Science Vol. 2, No. 3, pp. 109â118, 2013.
N. Chawla, N. Japkowicz and A. Kolcz, âEditorial: Special Issue on Learning from Imbalanced Data Setsâ, SIGKDD Explorations, Vol. 6, No. 1, pp. 1â6, 2004.
N. Chawla et al., âSMOTE: Synthetic Minority Over-Sampling Techniqueâ, Journal of Artificial Intelligence Research, Vol. 16, pp. 321â357, 2002.
N. Chawla et al., âData mining for imbalanced datasets: An overviewâ, in Data Mining and Knowledge Discovery Handbook, Springer, pp. 853â867, 2005.
B. X. Wang and N. Japkowicz, âImbalanced Data Set Learning with Synthetic Samplesâ, Proc. IRIS Machine Learning Workshop, 2004.
H. He, Y. Bai, E. A. Garcia, and S. Li, âADASYN: Adaptive synthetic sampling approach for imbalanced learningâ, Proc. IEEE Int. Joint Conf. Neural Netw., IEEE World Congr. Comput. Intell., pp. 1322â1328, 2008.
H. Han, W. Wang, and B. Mao, âBorderline-SMOTE: A new oversampling method in Imbalanced Data-sets Learningâ, In. ICIC 2005. LNCS, Vol. 3644, pp. 878â887, Springer, Heidelberg, 2005.
C. Bunkhumpornpat, K. Sinapiromsaran, and C. Lursinsap, âSafe-Level-SMOTE: Safe Level- Synthetic MI Over-Sampling Technique for handling the Class Imbalance Problemâ, PADD2009, LNAI, Vol. 5476, pp. 475â482, Springer, 2009.
J. Huang and C. X. Ling, âUsing AUC and Accuracy in Evaluating Learning Algorithmsâ, IEEE Transactions on Knowledge and Data Engineering, Vol. 17, No. 3, March 2005.
A. Gosain and S. Sardana, ââHandling Class Imbalance Problem Using Oversampling Techniques: A Reviewââ, communicated in International Conference on Advances in Computing, Communications and Informatics (ICACCI) 2017, Manipal, Karnataka, India, September 2017.
Buckland, M., Gey, F., âThe Relationship between Recall and Precisionâ, Journal of the American Society for Information Science 45(1), pp. 12â19, 1994.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
Âİ 2019 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Gosain, A., Sardana, S. (2019). Farthest SMOTE: A Modified SMOTE Approach. In: Behera, H., Nayak, J., Naik, B., Abraham, A. (eds) Computational Intelligence in Data Mining. Advances in Intelligent Systems and Computing, vol 711. Springer, Singapore. https://doi.org/10.1007/978-981-10-8055-5_28
Download citation
DOI: https://doi.org/10.1007/978-981-10-8055-5_28
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-10-8054-8
Online ISBN: 978-981-10-8055-5
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)