Scholarlink

<p>Data imputation is a technique for predicting suitable data for missing value healthcare datasets that contain patient and healthcare practitioner information. As medical data’s volume and complexity increase, precise data prediction is essential for informed decision-making and comprehensive analysis. This work introduces a novel method for pre-processing incomplete data, Class Center Mean Value Imputation, using the Enhanced Firefly Algorithm with Courtship Learning (CCMVI-EFA-CL). This method is distinguished by its precise estimations for Missing Completely At Random (MCAR) missingness. Using the Minkowski distance, our proposed method calculates the threshold for accurate data imputation for each class. This threshold is then optimized using an enhanced firefly algorithm with courtship learning. The method combines class center mean value imputation with an enhanced firefly algorithm and courtship learning within an iterative learning framework, ensuring accuracy and efficiency in handling medical datasets. Our proposed method has shown significant improvements in Mean Squared Error (MSE), Root Mean Squared Error (RMSE), R-squared (R<InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="10115_2025_2489_Article_IEq1.gif" Format="GIF" Height="10" Rendition="HTML" Resolution="72" Type="Linedraw" Width="8" /> </InlineMediaObject> <EquationSource Format="TEX">\(^2\)</EquationSource> <EquationSource Format="MATHML"><math> <mmultiscripts> <mrow /> <mrow /> <mn>2</mn> </mmultiscripts> </math></EquationSource> </InlineEquation>) and Mean Absolute Error (MAE). Experimental results from five benchmark healthcare datasets underscore the superior accuracy and robustness of the proposed method. By outperforming established methods such as Class Center Mean Value Imputation (CCMVI), Single Imputation Chained Equations (SICE), Multivariate Imputation by Chained Equations (MICE) and Class Center-based Firefly Algorithm (C3FA) in terms of accuracy, CCMVI-EFA-CL emerges as a promising solution for healthcare data management. The use of Minkowski distance for threshold calculation and its optimization using the enhanced firefly algorithm with courtship learning further enhances the accuracy and robustness of this method.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Accurate data imputation in healthcare with optimized class thresholds using enhanced firefly algorithm

  • 期刊论文

  • 发表日期：

    2025年6月14日

查看全文
  • Subhashish Nayak,
  • P. M. Khilar

摘要

Data imputation is a technique for predicting suitable data for missing value healthcare datasets that contain patient and healthcare practitioner information. As medical data’s volume and complexity increase, precise data prediction is essential for informed decision-making and comprehensive analysis. This work introduces a novel method for pre-processing incomplete data, Class Center Mean Value Imputation, using the Enhanced Firefly Algorithm with Courtship Learning (CCMVI-EFA-CL). This method is distinguished by its precise estimations for Missing Completely At Random (MCAR) missingness. Using the Minkowski distance, our proposed method calculates the threshold for accurate data imputation for each class. This threshold is then optimized using an enhanced firefly algorithm with courtship learning. The method combines class center mean value imputation with an enhanced firefly algorithm and courtship learning within an iterative learning framework, ensuring accuracy and efficiency in handling medical datasets. Our proposed method has shown significant improvements in Mean Squared Error (MSE), Root Mean Squared Error (RMSE), R-squared (R \(^2\) 2 ) and Mean Absolute Error (MAE). Experimental results from five benchmark healthcare datasets underscore the superior accuracy and robustness of the proposed method. By outperforming established methods such as Class Center Mean Value Imputation (CCMVI), Single Imputation Chained Equations (SICE), Multivariate Imputation by Chained Equations (MICE) and Class Center-based Firefly Algorithm (C3FA) in terms of accuracy, CCMVI-EFA-CL emerges as a promising solution for healthcare data management. The use of Minkowski distance for threshold calculation and its optimization using the enhanced firefly algorithm with courtship learning further enhances the accuracy and robustness of this method.