Advances In Algorithms For Re-Sampling Class-Imbalanced Educational Data Sets

Abstract

Real world data sets often contain disproportionate sample sizes of observed groups making it difficultfor predictive analytics algorithms. One of the many ways to combat inherent bias from class imbalancedata is to perform re-sampling. In this book chapter we discuss popular re-sampling methods proposedin research literature, such as Synthetic Minority Over-sampling Technique (SMOTE) and PropensityScore Matching (PSM). We provide an insight into recent advances and our own novel algorithms underthe umbrella term of Over-sampling Using Propensity Scores (OUPS). Using simulation we conductexperiments that result in statistical improvement in accuracy and sensitivity by using these new algorithmicapproaches.

Publication Date

12-12-2016

Publication Title

Artificial Intelligence: Concepts, Methodologies, Tools, and Applications

Volume

2

Number of Pages

1000-1030

Document Type

Article; Book Chapter

Personal Identifier

scopus

DOI Link

https://doi.org/10.4018/978-1-5225-1759-7.ch040

Socpus ID

85018566836 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/85018566836

This document is currently not available here.

Share

COinS