Global-and-Local Aware Data Generation for the Class Imbalance Problem

In many real-world classification applications such as fake news detection, the training data can be extremely imbalanced, which brings challenges to existing classifiers as the majority classes dominate the loss functions of classifiers. Oversampling techniques such as SMOTE are effective approaches to tackle the class imbalance problem by producing more synthetic minority samples. Despite their success, the majority of existing oversampling methods only consider local data distributions when generating minority samples, which can result in noisy minority samples that do not fit global data distributions or interleave with majority classes. Hence, in this paper, we study the class imbalance problem by simultaneously exploring local and global data information since: (i) the local data distribution could give detailed information for generating minority samples; and (ii) the global data distribution could provide guidance to avoid generating outliers or samples that interleave with majority classes. Specifically, we propose a novel framework GL-GAN, which leverages the SMOTE method to explore local distribution in a learned latent space and employs GAN to capture the global information, so that synthetic minority samples can be generated under even extremely imbalanced scenarios. Experimental results on diverse real data sets demonstrate the effectiveness of our GL-GAN framework in producing realistic and discriminative minority samples for improving the classification performance of various classifiers on imbalanced training data. Our code is available at https://github.com/wentao-repo/GL-GAN .

• Publication year

  2020

• Venue

  SDM

• Publication date

  2020-01-01

• Fields of study

  Computer Science

• Identifiers
  DOI 10.1137/1.9781611976236.35
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Smote-variants: A python implementation of 85 minority oversampling techniques

  2019cited by this paper
• Effective data generation for imbalanced learning using conditional generative adversarial networks

  2018cited by this paper
• BAGAN: Data Augmentation with Balancing GAN

  2018cited by this paper
• Synthetic Oversampling with the Majority Class: A New Perspective on Handling Extreme Imbalance

  2018cited by this paper
• GENERATIVE ADVERSARIAL NETS

  2018cited by this paper
• Triangle Generative Adversarial Networks

  2017cited by this paper
• Fake News Detection on Social Media: A Data Mining Perspective

  2017cited by this paper
• Noise Reduction A Priori Synthetic Over-Sampling for class imbalanced data sets

  2017cited by this paper
• Imbalanced-learn: A Python Toolbox to Tackle the Curse of Imbalanced Datasets in Machine Learning

  2016cited by this paper
• Learning from imbalanced data: open challenges and future directions

  2016cited by this paper
• To Combat Multi-Class Imbalanced Problems by Means of Over-Sampling Techniques

  2016cited by this paper
• Conditional Generative Adversarial Nets

  2014cited by this paper
• Deep learning in neural networks: An overview

  2014cited by this paper
• Class Imbalance Problem in Data Mining Review

  2013cited by this paper
• An overview of classification algorithms for imbalanced datasets

  2012influential reference
• Exploratory Undersampling for Class-Imbalance Learning

  2009cited by this paper
• Increasing the accuracy of neural network classification using refined training data

  2009cited by this paper
• ADASYN: Adaptive synthetic sampling approach for imbalanced learning

  2008cited by this paper
• Generative Oversampling for Mining Imbalanced Datasets

  2007cited by this paper
• Under-Sampling Approaches for Improving Prediction of the Minority Class in an Imbalanced Dataset

  2006cited by this paper
• Evaluation of neural networks and data mining methods on a credit assessment task for class imbalance problem

  2006cited by this paper
• Borderline-SMOTE: A New Over-Sampling Method in Imbalanced Data Sets Learning

  2005influential reference
• SPECIAL ISSUE ON LEARNING FROM IMBALANCED DATA SETS

  2004cited by this paper
• SMOTE: Synthetic Minority Over-sampling Technique

  2002influential reference
• Comparison of the predicted and observed secondary structure of T4 phage lysozyme.

  1975cited by this paper

• Least Information Spectral GAN With Time-Series Data Augmentation for Industrial IoT

  2025cites this paper
• Spatiotemporal deep learning models for detection of rapid intensification in cyclones

  2025cites this paper
• Fraud Detection Based on Graph Neural Network

  2024cites this paper
• A Lightweight Randomized Nonlinear Dictionary Learning Method using Random Vector Functional Link

  2024cites this paper
• Quality-Aware Self-Training on Differentiable Synthesis of Rare Relational Data

  2023influential citation
• Generative Diffusion Models on Graphs: Methods and Applications

  2023cites this paper
• Towards Automated Imbalanced Learning with Deep Hierarchical Reinforcement Learning

  2022cites this paper
• A novel textual data augmentation method for identifying comparative text from user-generated content

  2022cites this paper
• SOS: Score-based Oversampling for Tabular Data

  2022influential citation
• Semantic inpainting on segmentation map via multi-expansion loss

  2022cites this paper
• Learning from Imbalanced Crowdsourced Labeled Data

  2022cites this paper
• An empirical study on the joint impact of feature selection and data resampling on imbalance classification

  2021cites this paper
• Labeled Data Generation with Inexact Supervision

  2021cites this paper
• Pick and Choose: A GNN-based Imbalanced Learning Approach for Fraud Detection

  2021cites this paper
• Detecting Fake News Using Machine Learning : A Systematic Literature Review

  2021cites this paper
• Alike and Unlike: Resolving Class Imbalance Problem in Financial Credit Risk Assessment

  2020influential citation