A comparative study of missing value imputation with multiclass classification for clinical heart failure data

Yan Zhang; Chandra Kambhampati; Darryl Davis; Kevin Goode; John Cleland

doi:10.1109/FSKD.2012.6233805

A comparative study of missing value imputation with multiclass classification for clinical heart failure data

Yan Zhang, Chandra Kambhampati, Darryl Davis, Kevin Goode, John Cleland

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Clinical data often contains missing values.
Imputation is one of the best known schemes to overcome the
drawbacks associated with missing values in data mining tasks.
In this work, we compared several imputation methods and
analyzed their performance when applied to different
classification algorithms. A clinical heart failure data set was
used in these experiments. The results showed that there is no
universal imputation method that performs best for all
classifiers. Some imputation-classification combinations are
recommended for the processing of clinical heart failure data.

Original language	English
Title of host publication	2012 9th International Conference on Fuzzy Systems and Knowledge Discovery
Publisher	IEEE
Pages	2840 - 2844
ISBN (Electronic)	978-1-4673-0024-7
ISBN (Print)	978-1-4673-0025-4
DOIs	https://doi.org/10.1109/FSKD.2012.6233805
Publication status	Published - 2012

Keywords

missing value
imputation
classification
clinical data
heart failure

Documents and Links

10.1109/FSKD.2012.6233805

Fingerprint Dive into the research topics of 'A comparative study of missing value imputation with multiclass classification for clinical heart failure data'. Together they form a unique fingerprint.

Cite this

Zhang, Y., Kambhampati, C., Davis, D., Goode, K., & Cleland, J. (2012). A comparative study of missing value imputation with multiclass classification for clinical heart failure data. In 2012 9th International Conference on Fuzzy Systems and Knowledge Discovery (pp. 2840 - 2844). IEEE. https://doi.org/10.1109/FSKD.2012.6233805

@inproceedings{4545a0391f0a45e9b92e9e91114396e3,

title = "A comparative study of missing value imputation with multiclass classification for clinical heart failure data",

abstract = "Clinical data often contains missing values.Imputation is one of the best known schemes to overcome thedrawbacks associated with missing values in data mining tasks.In this work, we compared several imputation methods andanalyzed their performance when applied to differentclassification algorithms. A clinical heart failure data set wasused in these experiments. The results showed that there is nouniversal imputation method that performs best for allclassifiers. Some imputation-classification combinations arerecommended for the processing of clinical heart failure data. ",

keywords = "missing value, imputation, classification, clinical data, heart failure",

author = "Yan Zhang and Chandra Kambhampati and Darryl Davis and Kevin Goode and John Cleland",

year = "2012",

doi = "10.1109/FSKD.2012.6233805",

language = "English",

isbn = "978-1-4673-0025-4",

pages = "2840 -- 2844",

booktitle = "2012 9th International Conference on Fuzzy Systems and Knowledge Discovery",

publisher = "IEEE",

}

A comparative study of missing value imputation with multiclass classification for clinical heart failure data. / Zhang, Yan; Kambhampati, Chandra; Davis, Darryl; Goode, Kevin; Cleland, John.

2012 9th International Conference on Fuzzy Systems and Knowledge Discovery. IEEE, 2012. p. 2840 - 2844.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - A comparative study of missing value imputation with multiclass classification for clinical heart failure data

AU - Zhang, Yan

AU - Kambhampati, Chandra

AU - Davis, Darryl

AU - Goode, Kevin

AU - Cleland, John

PY - 2012

Y1 - 2012

N2 - Clinical data often contains missing values.Imputation is one of the best known schemes to overcome thedrawbacks associated with missing values in data mining tasks.In this work, we compared several imputation methods andanalyzed their performance when applied to differentclassification algorithms. A clinical heart failure data set wasused in these experiments. The results showed that there is nouniversal imputation method that performs best for allclassifiers. Some imputation-classification combinations arerecommended for the processing of clinical heart failure data.

AB - Clinical data often contains missing values.Imputation is one of the best known schemes to overcome thedrawbacks associated with missing values in data mining tasks.In this work, we compared several imputation methods andanalyzed their performance when applied to differentclassification algorithms. A clinical heart failure data set wasused in these experiments. The results showed that there is nouniversal imputation method that performs best for allclassifiers. Some imputation-classification combinations arerecommended for the processing of clinical heart failure data.

KW - missing value

KW - imputation

KW - classification

KW - clinical data

KW - heart failure

U2 - 10.1109/FSKD.2012.6233805

DO - 10.1109/FSKD.2012.6233805

M3 - Conference contribution

SN - 978-1-4673-0025-4

SP - 2840

EP - 2844

BT - 2012 9th International Conference on Fuzzy Systems and Knowledge Discovery

PB - IEEE

ER -