Scholarlink

In medical research and diagnostics, missing data can severely impact the reliability of analyses and predictions. Accurate data imputation is critical for maintaining the quality and usability of datasets. Traditional imputation methods often struggle with complex and varied data structures, leading to suboptimal outcomes. To address this challenge, this study introduces Multi-Model Imputation (MMMI), a novel approach that integrates multiple machine learning (ML) techniques to enhance data imputation accuracy. MMMI combines K-Nearest Neighbors (KNN) and Random Forest (RF) algorithms for audiogram data imputation. By leveraging KNN’s local relationship modeling, and RF’s ability to handle complex interactions MMMI improves both accuracy and consistency in data imputation. Comparative analyses reveal that MMMI outperforms traditional imputation methods, demonstrating a 2–20% improvement over KNN and RF in audiogram predictions. These results underscore MMMI’s effectiveness and its potential for broader applications in medical data imputation, leading to higher-quality and more reliable datasets.

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

Methodology of Multi-model Imputation: Novel Imputation Method Using Multiple Known Models

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Sarah Beaver,
  • Yashu Vashishath,
  • Renee Bryce

摘要

In medical research and diagnostics, missing data can severely impact the reliability of analyses and predictions. Accurate data imputation is critical for maintaining the quality and usability of datasets. Traditional imputation methods often struggle with complex and varied data structures, leading to suboptimal outcomes. To address this challenge, this study introduces Multi-Model Imputation (MMMI), a novel approach that integrates multiple machine learning (ML) techniques to enhance data imputation accuracy. MMMI combines K-Nearest Neighbors (KNN) and Random Forest (RF) algorithms for audiogram data imputation. By leveraging KNN’s local relationship modeling, and RF’s ability to handle complex interactions MMMI improves both accuracy and consistency in data imputation. Comparative analyses reveal that MMMI outperforms traditional imputation methods, demonstrating a 2–20% improvement over KNN and RF in audiogram predictions. These results underscore MMMI’s effectiveness and its potential for broader applications in medical data imputation, leading to higher-quality and more reliable datasets.