Skew Probabilistic Neural Networks for Learning from Imbalanced Data

Real-world datasets often exhibit imbalanced data distribution, where certain class levels are severely underrepresented. In such cases, traditional pattern classifiers have shown a bias towards the majority class, impeding accurate predictions for the minority class. This paper introduces an imbalanced data-oriented approach using probabilistic neural networks (PNNs) with a skew normal probability kernel to address this major challenge. PNNs are known for providing probabilistic outputs, enabling quantification of prediction confidence and uncertainty handling. By leveraging the skew normal distribution, which offers increased flexibility, particularly for imbalanced and non-symmetric data, our proposed Skew Probabilistic Neural Networks (SkewPNNs) can better represent underlying class densities. To optimize the performance of the proposed approach on imbalanced datasets, hyperparameter fine-tuning is imperative. To this end, we employ a population-based heuristic algorithm, Bat optimization algorithms, for effectively exploring the hyperparameter space. We also prove the statistical consistency of the density estimates which suggests that the true distribution will be approached smoothly as the sample size increases. Experimental simulations have been conducted on different synthetic datasets, comparing various benchmark-imbalanced learners. Our real-data analysis shows that SkewPNNs substantially outperform state-of-the-art machine learning methods for both balanced and imbalanced datasets in most experimental settings.

[1]  Shraddha M. Naik,et al.  Ten years of generative adversarial nets (GANs): a survey of the state-of-the-art , 2023, Mach. Learn. Sci. Technol..

[2]  L. xilinx Wang,et al.  A Hybrid Evolutionary Under-sampling Method for Handling the Class Imbalance Problem with Overlap in Credit Classification , 2022, Journal of Systems Science and Systems Engineering.

[3]  Qinghua Gu,et al.  A novel Random Forest integrated model for imbalanced data classification problem , 2022, Knowl. Based Syst..

[4]  Yotam Elor,et al.  To SMOTE, or not to SMOTE? , 2022, ArXiv.

[5]  Qi Han,et al.  An Improved Unbalanced Data Classification Method Based on Hybrid Sampling Approach , 2021, 2021 IEEE 4th International Conference on Big Data and Artificial Intelligence (BDAI).

[6]  Hao Wang,et al.  Delving into Deep Imbalanced Regression , 2021, ICML.

[7]  Z Rustam,et al.  Preprocessing Unbalanced Data using Support Vector Machine with Method K-Nearest Neighbors for Cerebral Infarction Classification , 2021, Journal of Physics: Conference Series.

[8]  Jakub Klikowski,et al.  Hellinger Distance Weighted Ensemble for Imbalanced Data Stream Classification , 2021, J. Comput. Sci..

[9]  Tanujit Chakraborty,et al.  Hellinger Net: A Hybrid Imbalance Learning Model to Improve Software Defect Prediction , 2020, IEEE Transactions on Reliability.

[10]  Radhouane Guermazi,et al.  Enhancing techniques for learning decision trees from imbalanced data , 2020, Adv. Data Anal. Classif..

[11]  D. Dunson,et al.  Classification Trees for Imbalanced Data: Surface-to-Volume Regularization , 2020, Journal of the American Statistical Association.

[12]  Michal Koziarski,et al.  CSMOUTE: Combined Synthetic Oversampling and Undersampling Technique for Imbalanced Data Classification , 2020, 2021 International Joint Conference on Neural Networks (IJCNN).

[13]  Bart Baesens,et al.  robROSE: A robust approach for dealing with imbalanced data in fraud detection , 2020, Statistical Methods & Applications.

[14]  YONGQING ZHANG,et al.  Evolutionary-Based Ensemble Under-Sampling for Imbalanced Data , 2019, 2019 16th International Computer Conference on Wavelet Active Media Technology and Information Processing.

[15]  Aman Jantan,et al.  Comprehensive Review of Artificial Neural Network Applications to Pattern Recognition , 2019, IEEE Access.

[16]  Shraddha M. Naik,et al.  Bat algorithm-based weighted Laplacian probabilistic neural network , 2019, Neural Computing and Applications.

[17]  Suzhen Wang,et al.  Imbalance-XGBoost: Leveraging Weighted and Focal Losses for Binary Label-Imbalanced Classification with XGBoost , 2019, Pattern Recognit. Lett..

[18]  Tanujit Chakraborty,et al.  Superensemble classifier for improving predictions in imbalanced datasets , 2018, Communications in Statistics: Case Studies, Data Analysis and Applications.

[19]  Francisco Herrera,et al.  Learning from Imbalanced Data Sets , 2018, Springer International Publishing.

[20]  Olivier Bachem,et al.  Assessing Generative Models via Precision and Recall , 2018, NeurIPS.

[21]  Adelchi Azzalini,et al.  The Skew-Normal and Related Families , 2018 .

[22]  Chen Zhang,et al.  Interaction between BDNF and TNF-α genes in schizophrenia , 2018, Psychoneuroendocrinology.

[23]  Fatemeh Afsari,et al.  Hesitant fuzzy decision tree approach for highly imbalanced data classification , 2017, Appl. Soft Comput..

[24]  Dimitris N. Metaxas,et al.  Addressing Imbalance in Multi-Label Classification Using Structured Hellinger Forests , 2017, AAAI.

[25]  Fernando Nogueira,et al.  Imbalanced-learn: A Python Toolbox to Tackle the Curse of Imbalanced Datasets in Machine Learning , 2016, J. Mach. Learn. Res..

[26]  Bartosz Krawczyk,et al.  Learning from imbalanced data: open challenges and future directions , 2016, Progress in Artificial Intelligence.

[27]  Krung Sinapiromsaran,et al.  Decision tree induction based on minority entropy for the class imbalance problem , 2016, Pattern Analysis and Applications.

[28]  Jian Wang,et al.  Improved probabilistic neural networks with self-adaptive strategies for transformer fault diagnosis problem , 2016 .

[29]  Erwan Scornet,et al.  A random forest guided tour , 2015, TEST.

[30]  Swagatam Das,et al.  Near-Bayesian Support Vector Machines for imbalanced data classification with equal or unequal misclassification costs , 2015, Neural Networks.

[31]  Debasis Kundu,et al.  Geometric Skew Normal Distribution , 2014, Sankhya B.

[32]  Joshua D. Knowles,et al.  Hellinger Distance Trees for Imbalanced Streams , 2014, 2014 22nd International Conference on Pattern Recognition.

[33]  K. Murase,et al.  MWMOTE--Majority Weighted Minority Oversampling Technique for Imbalanced Data Set Learning , 2014, IEEE Transactions on Knowledge and Data Engineering.

[34]  Rok Blagus,et al.  SMOTE for high-dimensional class-imbalanced data , 2013, BMC Bioinformatics.

[35]  Sebastián Ventura,et al.  Weighted Data Gravitation Classification for Standard and Imbalanced Data , 2013, IEEE Transactions on Cybernetics.

[36]  Josef Kittler,et al.  Inverse random under sampling for class imbalance problem and its application to multi-label classification , 2012, Pattern Recognit..

[37]  M. A. H. Farquad,et al.  Preprocessing unbalanced data using support vector machine , 2012, Decis. Support Syst..

[38]  Sotiris B. Kotsiantis,et al.  Decision trees: a recent overview , 2011, Artificial Intelligence Review.

[39]  Xin-She Yang,et al.  A New Metaheuristic Bat-Inspired Algorithm , 2010, NICSO.

[40]  A. Azzalini,et al.  Statistical applications of the multivariate skew normal distribution , 2009, 0911.2093.

[41]  Yue-Shi Lee,et al.  Cluster-based under-sampling approaches for imbalanced data distributions , 2009, Expert Syst. Appl..

[42]  David A. Cieslak,et al.  Learning Decision Trees for Unbalanced Data , 2008, ECML/PKDD.

[43]  Haibo He,et al.  ADASYN: Adaptive synthetic sampling approach for imbalanced learning , 2008, 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence).

[44]  Markus Neuhäuser,et al.  Wilcoxon Signed Rank Test , 2006 .

[45]  Hui Han,et al.  Borderline-SMOTE: A New Over-Sampling Method in Imbalanced Data Sets Learning , 2005, ICIC.

[46]  A. Azzalini The Skew‐normal Distribution and Related Multivariate Families * , 2005 .

[47]  M. Genton Discussion of ‘‘The Skew‐normal’’ , 2005 .

[48]  Stephen Kwek,et al.  Applying Support Vector Machines to Imbalanced Datasets , 2004, ECML.

[49]  Jose Almer T. Sanqui,et al.  Characterization of the skew-normal distribution , 2004 .

[50]  Arjun K. Gupta,et al.  A multivariate skew normal distribution , 2004 .

[51]  Nitesh V. Chawla,et al.  SMOTEBoost: Improving Prediction of the Minority Class in Boosting , 2003, PKDD.

[52]  Guoqiang Peter Zhang,et al.  Neural networks for classification: a survey , 2000, IEEE Trans. Syst. Man Cybern. Part C.

[53]  Wee Ser,et al.  Probabilistic neural-network structure determination for pattern classification , 2000, IEEE Trans. Neural Networks Learn. Syst..

[54]  Brian D. Ripley,et al.  Neural Networks and Related Methods for Classification , 1994 .

[55]  David J. Montana,et al.  A Weighted Probabilistic Neural Network , 1991, NIPS.

[56]  Donald F. Specht,et al.  Probabilistic neural networks and the polynomial Adaline as complementary techniques for classification , 1990, IEEE Trans. Neural Networks.

[57]  J. Hanley,et al.  The meaning and use of the area under a receiver operating characteristic (ROC) curve. , 1982, Radiology.

[58]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[59]  Dennis L. Wilson,et al.  Asymptotic Properties of Nearest Neighbor Rules Using Edited Data , 1972, IEEE Trans. Syst. Man Cybern..

[60]  E. Parzen On Estimation of a Probability Density Function and Mode , 1962 .

[61]  Taochun Wang,et al.  An automatic sampling ratio detection method based on genetic algorithm for imbalanced data classification , 2021, Knowl. Based Syst..

[62]  Seetha Hari,et al.  Learning From Imbalanced Data , 2019, Advances in Computer and Electrical Engineering.

[63]  Huanyu Dong,et al.  BA-PNN-based methods for power transformer fault diagnosis , 2019, Adv. Eng. Informatics.

[64]  Francisco Herrera,et al.  SMOTE for Learning from Imbalanced Data: Progress and Challenges, Marking the 15-year Anniversary , 2018, J. Artif. Intell. Res..

[65]  Yiguang Liu,et al.  Improving Random Forest and Rotation Forest for highly imbalanced datasets , 2015, Intell. Data Anal..

[66]  David A. Cieslak,et al.  Hellinger distance decision trees are robust and skew-insensitive , 2011, Data Mining and Knowledge Discovery.

[67]  Trevor Hastie,et al.  Multi-class AdaBoost ∗ , 2009 .

[68]  B. Arnold,et al.  Characterizations of the skew-normal and generalized chi distributions , 2004 .

[69]  Richard Lippmann,et al.  Neural Network Classifiers Estimate Bayesian a posteriori Probabilities , 1991, Neural Computation.

[70]  D. Specht Probabilistic neural networks , 1990, Neural Networks.

[71]  A. Azzalini A class of distributions which includes the normal ones , 1985 .