Highly Imbalanced Classification of Gout Using Data Resampling and Ensemble Method

• DOI: 10.3390/a17030122
• Publication Date: 2024-03-15T16:02:39Z
• Authors: Xiaonan Si; Lei Wang; Wenchang Xu; Biao Wang; Wenbo Cheng
• Keywords: Resampling; Ensemble Learning

Gout is one of the most painful diseases in world. Accurate classification gout crucial for diagnosis and treatment which can potentially save lives. However, current methods classifying periods have demonstrated poor performance received little attention. This due to a significant data imbalance problem that affects learning attention majority minority classes. To overcome this problem, resampling method called ENaNSMOTE-Tomek link proposed. It uses extended natural neighbors generate samples fall within class then applies Tomek technique eliminate instances contribute noise. The model combines ensemble ’bagging’ with proposed improve quality generated samples. individual classifiers hybrid models on an imbalanced dataset taken from electronic medical records hospital evaluated. results demonstrate strategy more accurate than some techniques, accuracy 80.87% AUC 87.10%. indicates algorithm alleviate problems caused by help experts better diagnose their patients.