What question did this study set out to answer?

Review machine learning techniques for handling missing data and evaluate the imputation accuracy of k-nearest neighbor and missForest algorithms under varied missingness conditions.

October 27, 2021SHILAP Revista de lepidopterologíaOpen Access

A survey on missing data in machine learning

View Full Paper

Ask AI

Bookmark

Authors

TETlamelo Emmanuel TMThabiso Maupong DMDimane Mpoeleng

Discussion

Member takes

Overview

Comparative evaluation demonstrates robust data imputation in benchmark and industrial datasets, highlighting machine learning efficacy over record deletion.

Key Points

Review machine learning techniques for handling missing data and evaluate the imputation accuracy of k-nearest neighbor and missForest algorithms under varied missingness conditions.
Synthesized literature on machine learning imputation mechanisms, data applicability, operational constraints, and mechanisms of missingness (MCAR, MAR, MNAR).
Evaluated k-nearest neighbor (KNN) and missForest (random forest-based iterative imputation) across the Iris benchmark and a novel power plant fan dataset.
Introduced synthetic missing values into the test datasets at missingness rates ranging from 5% to 20%.
Both missForest and k-nearest neighbor algorithms successfully imputed missing values across datasets with induced missingness rates of 5% to 20%.
Iterative random forest and neighbor-based approaches effectively preserved dataset utility without requiring the omission of incomplete records.

Cite This Study

Emmanuel et al. (2021) studied this question.

synapsesocial.com/papers/69d7cff205ee2ba81dbee1ee https://doi.org/10.1186/s40537-021-00516-9

View Full Paper

Ask AI

Bookmark

October 27, 2021SHILAP Revista de lepidopterologíaOpen Access

A survey on missing data in machine learning

View Full Paper

Ask AI

Bookmark

Authors

TETlamelo Emmanuel TMThabiso Maupong DMDimane Mpoeleng

Discussion

Member takes

Overview

Comparative evaluation demonstrates robust data imputation in benchmark and industrial datasets, highlighting machine learning efficacy over record deletion.

Key Points

Review machine learning techniques for handling missing data and evaluate the imputation accuracy of k-nearest neighbor and missForest algorithms under varied missingness conditions.
Synthesized literature on machine learning imputation mechanisms, data applicability, operational constraints, and mechanisms of missingness (MCAR, MAR, MNAR).
Evaluated k-nearest neighbor (KNN) and missForest (random forest-based iterative imputation) across the Iris benchmark and a novel power plant fan dataset.
Introduced synthetic missing values into the test datasets at missingness rates ranging from 5% to 20%.
Both missForest and k-nearest neighbor algorithms successfully imputed missing values across datasets with induced missingness rates of 5% to 20%.
Iterative random forest and neighbor-based approaches effectively preserved dataset utility without requiring the omission of incomplete records.

Cite This Study

Emmanuel et al. (2021) studied this question.

synapsesocial.com/papers/69d7cff205ee2ba81dbee1ee https://doi.org/10.1186/s40537-021-00516-9

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark