Skip to main navigation Skip to search Skip to main content

Improving prediction accuracy for logistic regression on imbalanced datasets

  • Hao Zhang
  • , Zhuolin Li
  • , Hossain Shahriar
  • , Lixin Tao
  • , Prabir Bhattacharya
  • , Ying Qian
  • Kennesaw State University
  • Pace University
  • University at Albany

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

An imbalanced dataset is a dataset that has a majority class which is a class has far more example distributions than other classes. It is difficult to deal with unbalanced datasets in classification problems, and many classification algorithms do not perform well in unbalanced datasets. In this paper, we present our logistic regression analysis with Python on imbalanced datasets and determine different thresholds for classification according to the data proportion of imbalanced datasets.

Original languageEnglish
Title of host publicationProceedings - 2019 IEEE 43rd Annual Computer Software and Applications Conference, COMPSAC 2019
EditorsVladimir Getov, Jean-Luc Gaudiot, Nariyoshi Yamai, Stelvio Cimato, Morris Chang, Yuuichi Teranishi, Ji-Jiang Yang, Hong Va Leong, Hossian Shahriar, Michiharu Takemoto, Dave Towey, Hiroki Takakura, Atilla Elci, Susumu Takeuchi, Satish Puri
PublisherIEEE Computer Society
Pages918-919
Number of pages2
ISBN (Electronic)9781728126074
DOIs
StatePublished - Jul 2019
Event43rd Annual IEEE Computer Software and Applications Conference, COMPSAC 2019 - Milwaukee, United States
Duration: 15 Jul 2019 → 19 Jul 2019

Publication series

NameProceedings - International Computer Software and Applications Conference
Volume1
ISSN (Print)0730-3157

Conference

Conference43rd Annual IEEE Computer Software and Applications Conference, COMPSAC 2019
Country/TerritoryUnited States
CityMilwaukee
Period15/07/19 → 19/07/19

Keywords

  • Balanced-accuracy
  • Classification
  • Imbalanced-dataset
  • Logistic-regression
  • Threshold

Fingerprint

Dive into the research topics of 'Improving prediction accuracy for logistic regression on imbalanced datasets'. Together they form a unique fingerprint.

Cite this