Weighted k‐nearest neighbor based data complexity metrics for imbalanced datasets. (2nd June 2020)

• Record Type: Journal Article
• Authors: Singh, Deepika; Gosain, Anjana; Saha, Anju
• Is Part Of: Statistical analysis and data mining. Volume 13:Number 4(2020)
• Page Start: 394
• Page End: 404
• Publication Date: 2020-06-02
• Subjects: Bayes error -- class imbalance -- classification -- complexity metrics -- data complexity -- data‐level algorithms -- imbalance ratio -- overlapping -- small disjuncts -- undersampled; Data mining -- Statistical methods -- Periodicals; 006.312
• DOI: 10.1002/sam.11463
• Languages: English
• ISSNs: 1932-1864

Abstract

Abstract: Empirical behavior of a classifier depends strongly on the characteristics of the underlying imbalanced dataset; therefore, an analysis of intrinsic data complexity would appear to be vital in order to choose classifiers suitable for particular problems. Data complexity metrics (CMs), a fairly recent proposal, identify dataset features which imply some difficulty for the classification task and identify relationships with classifier accuracy. In this paper, we introduce two CMs for imbalanced datasets, which help in explaining the factors responsible for the deterioration in classifier performance. These metrics are based on the weighted k ‐nearest neighbors approach. The experiments are performed in MATLAB software using 48 simulated datasets and 22 real‐world datasets for different choices of neighborhood size k considered as 3, 5, 7, 9, 11. The results help to illustrate the usefulness of the proposed metrics.