Conceptual metaphor quantum correlation and radial basis extreme learning for predicting chronic kidney disease

•CMQMC-RBEL method combines structured and unstructured data for accurate CKD diagnosis.•Quantum mutual information technique identifies highly correlated features, enhancing detection precision.•Conceptual metaphor process refines unstructured Twitter health data to yield relevant terms for CKD analysis. Chronic kidney disease is a progressive condition that often remains unnoticed until substantial kidney damage has occurred, leading to severe complications like nerve damage, pregnancy issues, and potentially fatal kidney failure. Early diagnosis of chronic kidney disease is challenging due to a lack of distinct indicators. To enhance detection, a new machine learning technique called conceptual metaphor quantum mutual correlation and radial basis extreme learning is developed for chronic kidney disease diagnosis. This approach starts by extracting critical terms from unstructured datasets. It then applies quantum mutual information-based feature selection to identify highly correlated feature subsets from both unstructured and structured data. Finally, radial basis extreme learning is used for classification, overcoming imbalanced data and computational issues typical in extreme learning machines. The proposed method demonstrates remarkable accuracy (98 %), precision (88 %), and recall (92 %), surpassing traditional models, and holds significant potential for improving chronic kidney disease detection and treatment. Graphical representation of the proposed CMQMC-RBEL for CKD diagnosis. [Display omitted]