[論文レビュー] A Review of Machine Learning Techniques in Imbalanced Data and Future Trends

Elaheh Jafarigol, Theodore B. Trafali̇s+1 more|arXiv (Cornell University)|Oct 11, 2023

Imbalanced Data Classification Techniques被引用数 8

ひとこと要約

This paper surveys techniques for imbalanced data by reviewing 258 peer-reviewed works and provides structured guidance and future trends for researchers and practitioners.

ABSTRACT

For over two decades, detecting rare events has been a challenging task among researchers in the data mining and machine learning domain. Real-life problems inspire researchers to navigate and further improve data processing and algorithmic approaches to achieve effective and computationally efficient methods for imbalanced learning. In this paper, we have collected and reviewed 258 peer-reviewed papers from archival journals and conference papers in an attempt to provide an in-depth review of various approaches in imbalanced learning from technical and application perspectives. This work aims to provide a structured review of methods used to address the problem of imbalanced data in various domains and create a general guideline for researchers in academia or industry who want to dive into the broad field of machine learning using large-scale imbalanced data.

研究の動機と目的

• Motivate the study by addressing the challenge of detecting rare events in real-life problems.
• Provide an in-depth, structured review of methods used to address imbalanced data across domains.
• Synthesize findings to create general guidelines for researchers in academia or industry working with large-scale imbalanced data.

提案手法

• Collected and reviewed 258 peer-reviewed papers from archival journals and conference papers.
• Structured the review to cover technical and application perspectives of imbalanced learning.
• Aimed to create a general guideline for applying large-scale imbalanced data techniques in ML workflows.

実験結果

リサーチクエスチョン

• RQ1 What are the main methodological approaches used to tackle imbalanced data across different domains?
• RQ2 How do these imbalanced learning techniques perform in various application contexts?
• RQ3 What guidelines can be proposed to guide future research and practice in imbalanced data scenarios?

主な発見

• The literature encompasses a wide range of techniques addressing imbalanced data in both technical and application contexts.
• A structured review can help unify approaches and clarify best practices for imbalanced learning.
• The study aims to generate a general guideline to assist academia and industry in adopting imbalanced data methods.