BAMTGAN: A Balanced Augmentation Technique for Tabular Data

Jueun Jeong, Hanseok Jeong, Han Joon Kim

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

This paper presents BAMTGAN, a novel data augmentation technique that addresses the class imbalance problem and prevents mode collapse by utilizing a modified DCGAN model and a new similarity loss to generate diverse and realistic tabular data. BAMTGAN encodes each column to produce a feature map for each record, which is then converted back to its original tabular form an intermediate image format. Experimental results demonstrate that BAMTGAN provides a more substantial improvement in developing high-quality predictive models than existing augmentation methods.

Original languageEnglish
Title of host publication2023 9th International Conference on Applied System Innovation, ICASI 2023
EditorsShoou-Jinn Chang, Sheng-Joue Young, Artde Donald Kin-Tak Lam, Liang-Wen Ji, Stephen D. Prior
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages205-207
Number of pages3
ISBN (Electronic)9798350398380
DOIs
StatePublished - 2023
Event9th International Conference on Applied System Innovation, ICASI 2023 - Chiba, Japan
Duration: 21 Apr 202325 Apr 2023

Publication series

Name2023 9th International Conference on Applied System Innovation, ICASI 2023

Conference

Conference9th International Conference on Applied System Innovation, ICASI 2023
Country/TerritoryJapan
CityChiba
Period21/04/2325/04/23

Keywords

  • Data Augmentation
  • Deep Learning
  • GAN
  • Tabular data
  • Training data

Fingerprint

Dive into the research topics of 'BAMTGAN: A Balanced Augmentation Technique for Tabular Data'. Together they form a unique fingerprint.

Cite this