Utilizing Nearest-Neighbor Clustering for Addressing Imbalanced Datasets in Bioengineering

TL;DR

This paper introduces a new clustering method to improve classification accuracy in imbalanced datasets, particularly useful in medical diagnosis and similar fields.

Contribution

The novel LBNN algorithm uses KMOR for outlier detection and improves performance on imbalanced data.

Findings

The LBNN algorithm outperforms traditional models in precision, recall, and G-means on imbalanced datasets.

Experiments on KEEL datasets and real medical data confirm the effectiveness of the proposed method.

Replacing the inter-quartile range with KMOR improves outlier identification in one-class problems.

Abstract

Imbalance classification is common in scenarios like fault diagnosis, intrusion detection, and medical diagnosis, where obtaining abnormal data is difficult. This article addresses a one-class problem, implementing and refining the One-Class Nearest-Neighbor (OCNN) algorithm. The original inter-quartile range mechanism is replaced with the K-means with outlier removal (KMOR) algorithm for efficient outlier identification in the target class. Parameters are optimized by treating these outliers as non-target-class samples. A new algorithm, the Location-based Nearest-Neighbor (LBNN) algorithm, clusters one-class training data using KMOR and calculates the farthest distance and percentile for each test data point to determine if it belongs to the target class. Experiments cover parameter studies, validation on eight standard imbalanced datasets from KEEL, and three applications on real medical imbalanced datasets. Results show superior performance in precision, recall, and G-means compared to traditional classification models, making it effective for handling imbalanced data challenges.

Figures