A Robust Missing Value Imputation Method MifImpute For Incomplete Molecular Descriptor Data And Comparative Analysis With Other Missing Value Imputation Methods

TL;DR

This paper introduces MiFoImpute, a robust random forest-based method for imputing missing values in molecular descriptor data, demonstrating superior accuracy and efficiency over existing methods in pharmaceutical datasets.

Contribution

The paper presents MiFoImpute, a novel iterative imputation technique based on random forests that effectively handles high-dimensional molecular data with missing values.

Findings

MiFoImpute outperforms ten benchmark imputation methods in accuracy.

It maintains robustness across missing data rates from 10% to 30%.

The method is computationally efficient for high-dimensional datasets.

Abstract

Missing data imputation is an important research topic in data mining. Large-scale Molecular descriptor data may contains missing values (MVs). However, some methods for downstream analyses, including some prediction tools, require a complete descriptor data matrix. We propose and evaluate an iterative imputation method MiFoImpute based on a random forest. By averaging over many unpruned regression trees, random forest intrinsically constitutes a multiple imputation scheme. Using the NRMSE and NMAE estimates of random forest, we are able to estimate the imputation error. Evaluation is performed on two molecular descriptor datasets generated from a diverse selection of pharmaceutical fields with artificially introduced missing values ranging from 10% to 30%. The experimental result demonstrates that missing values has a great impact on the effectiveness of imputation techniques and our method MiFoImpute is more robust to missing value than the other ten imputation methods used as benchmark. Additionally, MiFoImpute exhibits attractive computational efficiency and can cope with high-dimensional data.