期刊
EXPERIMENTAL BIOLOGY AND MEDICINE
卷 247, 期 22, 页码 1972-1987出版社
SAGE PUBLICATIONS LTD
DOI: 10.1177/15353702221121602
关键词
Imputation; missing data; deep learning; model checking; evaluation metrics
资金
- Natural Sciences and Engineering Research Council of Canada
- Southern Ontario Smart Computing Innovation Platform
- Canadian Department of National Defense: Innovation for Defense Excellence & Security Program
This article investigates the limitations of evaluating deep learning-based imputation models using alternative metrics, and proposes a new aggregated metric and evaluation methodology to assess the reconstruction performance of the models.
There is growing interest in imputing missing data in tabular datasets using deep learning. Existing deep learning-based imputation models have been commonly evaluated using root mean square error (RMSE) as the predictive accuracy metric. In this article, we investigate the limitations of assessing deep learning-based imputation models by conducting a comparative analysis between RMSE and alternative metrics in the statistical literature including qualitative, predictive accuracy, statistical distance, and descriptive statistics. We design a new aggregated metric, called reconstruction loss (RL), to evaluate deep learning-based imputation models. We also develop and evaluate a novel imputation evaluation methodology based on RL. To minimize model and dataset biases, we use a regression imputation model and two different deep learning imputation models: denoising autoencoders and generative adversarial nets. We also use two tabular datasets from different industry sectors: health care and financial. Our results show that the proposed methodology is effective in evaluating multiple properties of the deep learning-based imputation model's reconstruction performance.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据