Oversampling for Imbalanced Learning Based on K-Means and SMOTE

Abstract

Learning from class-imbalanced data continues to be a common and challenging problem in supervised learning as standard classification algorithms are designed to handle balanced class distributions. While different strategies exist to tackle this problem, methods which generate artificial data to achieve a balanced class distribution are more versatile than modifications to the classification algorithm. Such techniques, called oversamplers, modify the training data, allowing any classifier to be used with class-imbalanced datasets. Many algorithms have been proposed for this task, but most are complex and tend to generate unnecessary noise. This work presents a simple and effective oversampling method based on k-means clustering and SMOTE oversampling, which avoids the generation of noise and effectively overcomes imbalances between and within classes. Empirical results of extensive experiments with 71 datasets show that training data oversampled with the proposed method improves classification results. Moreover, k-means SMOTE consistently outperforms other popular oversampling methods. An implementation is made available in the python programming language.

arXiv comment19 pages, 8 figures

References in corpus (3)

• Scikit-learn: Machine Learning in Python
• SMOTE: Synthetic Minority Over-sampling Technique
• Imbalanced-learn: A Python Toolbox to Tackle the Curse of Imbalanced Datasets in Machine Learning

Cited by in corpus (24)

• Julia Language in Machine Learning: Algorithms, Applications, and Open Issues
• Weakly Supervised-Based Oversampling for High Imbalance and High Dimensionality Data Classification
• Minority Oversampling for Imbalanced Time Series Classification
• Comparing Multi-class, Binary and Hierarchical Machine Learning Classification schemes for variable stars
• Handling Imbalanced Datasets Through Optimum-Path Forest
• Multiwavelength Spectral Analysis and Neural Network Classification of Counterparts to 4FGL Unassociated Sources
• Fair Oversampling Technique using Heterogeneous Clusters
• X-ray Spectra and Multiwavelength Machine Learning Classification for Likely Counterparts to Fermi 3FGL Unassociated Sources
• Improving SMOTE via Fusing Conditional VAE for Data-adaptive Noise Filtering
• A Novel Resampling Technique for Imbalanced Dataset Optimization
• Handling Imbalanced Data: A Case Study for Binary Class Problems
• Enhancing Metabolic Syndrome Prediction with Hybrid Data Balancing and Counterfactuals
• Reproducible Machine Learning-based Voice Pathology Detection: Introducing the Pitch Difference Feature
• Sampling - Variational Auto Encoder - Ensemble: In the Quest of Explainable Artificial Intelligence
• SetConv: A New Approach for Learning from Imbalanced Data
• DRAST -- A Deep Learning and AST Based Approach for Bug Localization
• Let's Predict Who Will Move to a New Job
• Investigating bankruptcy prediction models in the presence of extreme class imbalance and multiple stages of economy
• Prediction of the Facial Growth Direction is Challenging
• A Method for Handling Multi-class Imbalanced Data by Geometry based Information Sampling and Class Prioritized Synthetic Data Generation (GICaPS)
• GMOTE: Gaussian based minority oversampling technique for imbalanced classification adapting tail probability of outliers
• Envelope Imbalance Learning Algorithm based on Multilayer Fuzzy C-means Clustering and Minimum Interlayer discrepancy
• Generating Synthetic Data with Locally Estimated Distributions for Disclosure Control
• Probabilistic Diagnostic Tests for Degradation Problems in Supervised Learning