Error Localization for Erroneous Data: Continuous Data, Linear Constraints

Data gathered in surveys, questionnaires, and censuses often contain a significant proportion of errors. If each record that fails a set of constraints (edits) is to be corrected, a reasonable model is to find the smallest (cheapest) set of fields which can be changed to yield a passing record. The...

Full description

Saved in:
Bibliographic Details
Published inSIAM journal on scientific and statistical computing Vol. 9; no. 5; pp. 922 - 931
Main Authors Garfinkel, R. S., Kunnathur, A. S., Liepins, G. E.
Format Journal Article
LanguageEnglish
Published Philadelphia Society for Industrial and Applied Mathematics 01.09.1988
Subjects
Linear programming
Localization
Operations management
Questionnaires
Online AccessGet full text

Cover

More Information
Summary:Data gathered in surveys, questionnaires, and censuses often contain a significant proportion of errors. If each record that fails a set of constraints (edits) is to be corrected, a reasonable model is to find the smallest (cheapest) set of fields which can be changed to yield a passing record. The problem is solved for continuous data and linear constraints using both a cutting plane algorithm based on the set-covering approach and a heuristic based on the simplex method. Extensive computational results are given.
ISSN:0196-5204
1064-8275
2168-3417
1095-7197
DOI:10.1137/0909063