The effect of missing data on classification quality

Michael Feldman; Adir Even; Yisrael Parmet

The effect of missing data on classification quality

Michael Feldman, Adir Even, Yisrael Parmet

Department of Industrial Engineering and Management

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

Abstract

The field of data quality management has long recognized the negative impact of data quality defects on decision quality. In many decision scenarios, this negative impact can be largely attributed to the mediating role played by decision-support models - with defected data, the estimation of such a model becomes less reliable and, as a result, the likelihood of flawed decisions increases. Drawing on that argument, this study presents a methodology for assessing the impact of quality defects on the likelihood of flawed decisions. The methodology is first presented at a high level, and then extended for analyzing the impact of missing values on binary Linear Discriminant Analysis (LDA) classifiers. To conclude, we discuss possible directions for extensions and future directions.

Original language	English
Title of host publication	Proceedings of ICIQ 2012
Subtitle of host publication	17th International Conference on Information Quality
Editors	Laure Berti-Equille, Isabelle Comyn-Wattiau, Monica Scannapieco
Publisher	MIT
Pages	229-242
Number of pages	14
ISBN (Electronic)	9781627483964
State	Published - 1 Jan 2012
Event	17th International Conference on Information Quality, ICIQ 2012 - Paris, France Duration: 16 Nov 2012 → 17 Nov 2012

Conference

Conference	17th International Conference on Information Quality, ICIQ 2012
Country/Territory	France
City	Paris
Period	16/11/12 → 17/11/12

Keywords

Classification
Data Quality
Decision Making
Linear Discriminant Analysis
Missing Values

ASJC Scopus subject areas

Information Systems
Safety, Risk, Reliability and Quality

Cite this

@inproceedings{b31d6a85d8fc48e591a1ac440c9d777d,

title = "The effect of missing data on classification quality",

abstract = "The field of data quality management has long recognized the negative impact of data quality defects on decision quality. In many decision scenarios, this negative impact can be largely attributed to the mediating role played by decision-support models - with defected data, the estimation of such a model becomes less reliable and, as a result, the likelihood of flawed decisions increases. Drawing on that argument, this study presents a methodology for assessing the impact of quality defects on the likelihood of flawed decisions. The methodology is first presented at a high level, and then extended for analyzing the impact of missing values on binary Linear Discriminant Analysis (LDA) classifiers. To conclude, we discuss possible directions for extensions and future directions.",

keywords = "Classification, Data Quality, Decision Making, Linear Discriminant Analysis, Missing Values",

author = "Michael Feldman and Adir Even and Yisrael Parmet",

year = "2012",

month = jan,

day = "1",

language = "English",

pages = "229--242",

editor = "Laure Berti-Equille and Isabelle Comyn-Wattiau and Monica Scannapieco",

booktitle = "Proceedings of ICIQ 2012",

publisher = "MIT",

}

The effect of missing data on classification quality. / Feldman, Michael; Even, Adir ; Parmet, Yisrael.

Proceedings of ICIQ 2012: 17th International Conference on Information Quality. ed. / Laure Berti-Equille; Isabelle Comyn-Wattiau; Monica Scannapieco. MIT, 2012. p. 229-242.

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

TY - GEN

T1 - The effect of missing data on classification quality

AU - Feldman, Michael

AU - Even, Adir

AU - Parmet, Yisrael

PY - 2012/1/1

Y1 - 2012/1/1

N2 - The field of data quality management has long recognized the negative impact of data quality defects on decision quality. In many decision scenarios, this negative impact can be largely attributed to the mediating role played by decision-support models - with defected data, the estimation of such a model becomes less reliable and, as a result, the likelihood of flawed decisions increases. Drawing on that argument, this study presents a methodology for assessing the impact of quality defects on the likelihood of flawed decisions. The methodology is first presented at a high level, and then extended for analyzing the impact of missing values on binary Linear Discriminant Analysis (LDA) classifiers. To conclude, we discuss possible directions for extensions and future directions.

AB - The field of data quality management has long recognized the negative impact of data quality defects on decision quality. In many decision scenarios, this negative impact can be largely attributed to the mediating role played by decision-support models - with defected data, the estimation of such a model becomes less reliable and, as a result, the likelihood of flawed decisions increases. Drawing on that argument, this study presents a methodology for assessing the impact of quality defects on the likelihood of flawed decisions. The methodology is first presented at a high level, and then extended for analyzing the impact of missing values on binary Linear Discriminant Analysis (LDA) classifiers. To conclude, we discuss possible directions for extensions and future directions.

KW - Classification

KW - Data Quality

KW - Decision Making

KW - Linear Discriminant Analysis

KW - Missing Values

UR - http://www.scopus.com/inward/record.url?scp=85077880705&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85077880705

SP - 229

EP - 242

BT - Proceedings of ICIQ 2012

A2 - Berti-Equille, Laure

A2 - Comyn-Wattiau, Isabelle

A2 - Scannapieco, Monica

PB - MIT

T2 - 17th International Conference on Information Quality, ICIQ 2012

Y2 - 16 November 2012 through 17 November 2012

ER -

The effect of missing data on classification quality

Abstract

Conference

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this