Establishing strong imputation performance of a denoising autoencoder in a wide range of missing data problems

Research output: Contribution to journalArticle

Standard

Harvard

APA

CBE

MLA

Vancouver

Author

RIS

TY - JOUR

T1 - Establishing strong imputation performance of a denoising autoencoder in a wide range of missing data problems

AU - Abiri, Najmeh

AU - Linse, Björn

AU - Edén, Patrik

AU - Ohlsson, Mattias

PY - 2019/7/26

Y1 - 2019/7/26

N2 - Dealing with missing data in data analysis is inevitable. Although powerful imputation methods that address this problem exist, there is still much room for improvement. In this study, we examined single imputation based on deep autoencoders, motivated by the apparent success of deep learning to efficiently extract useful dataset features. We have developed a consistent framework for both training and imputation. Moreover, we benchmarked the results against state-of-the-art imputation methods on different data sizes and characteristics. The work was not limited to the one-type variable dataset; we also imputed missing data with multi-type variables, e.g., a combination of binary, categorical, and continuous attributes. To evaluate the imputation methods, we randomly corrupted the complete data, with varying degrees of corruption, and then compared the imputed and original values. In all experiments, the developed autoencoder obtained the smallest error for all ranges of initial data corruption.

AB - Dealing with missing data in data analysis is inevitable. Although powerful imputation methods that address this problem exist, there is still much room for improvement. In this study, we examined single imputation based on deep autoencoders, motivated by the apparent success of deep learning to efficiently extract useful dataset features. We have developed a consistent framework for both training and imputation. Moreover, we benchmarked the results against state-of-the-art imputation methods on different data sizes and characteristics. The work was not limited to the one-type variable dataset; we also imputed missing data with multi-type variables, e.g., a combination of binary, categorical, and continuous attributes. To evaluate the imputation methods, we randomly corrupted the complete data, with varying degrees of corruption, and then compared the imputed and original values. In all experiments, the developed autoencoder obtained the smallest error for all ranges of initial data corruption.

KW - Autoencoder

KW - Deep learning

KW - Imputation

KW - Missing data

UR - http://www.scopus.com/inward/record.url?scp=85069939556&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2019.07.065

DO - 10.1016/j.neucom.2019.07.065

M3 - Article

JO - Neurocomputing

T2 - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -