How to Detect Possible Additional Outliers: Case of Interval Uncertainty

Research output: Contribution to journalArticleResearchpeer review

Authors

Research Organisations

External Research Organisations

  • University of Texas at El Paso
View graph of relations

Details

Original languageEnglish
Pages (from-to)100-106
Number of pages7
JournalReliable Computing
Volume28
Publication statusPublished - Jun 2021

Abstract

In many practical situations, measurements are characterized by interval uncertainty - namely, based on each measurement result, the only information that we have about the actual value of the measured quantity is that this value belongs to some interval. If several such intervals - corresponding to measuring the same quantity - have an empty intersection, this means that at least one of the corresponding measurement results is an outlier, caused by a malfunction of the measuring instrument. From the purely mathematical viewpoint, if the intersection is non-empty, there is no reason to be suspicious. However, from the practical viewpoint, if the intersection is too narrow - i.e., almost empty - then we should also be suspicious, and mark this as an possible additional outlier case. In this paper, we describe a natural way to formalize this idea, and an algorithm for detecting such additional possible outliers.

Keywords

    interval uncertainty, outliers, probabilistic approach

ASJC Scopus subject areas

Cite this

How to Detect Possible Additional Outliers: Case of Interval Uncertainty. / Dbouk, Hani; Schön, Steffen; Neumann, Ingo et al.
In: Reliable Computing, Vol. 28, 06.2021, p. 100-106.

Research output: Contribution to journalArticleResearchpeer review

Dbouk, H, Schön, S, Neumann, I & Kreinovichy, V 2021, 'How to Detect Possible Additional Outliers: Case of Interval Uncertainty', Reliable Computing, vol. 28, pp. 100-106. <https://www.cs.utep.edu/vladik/2020/tr20-67b.pdf>
Dbouk H, Schön S, Neumann I, Kreinovichy V. How to Detect Possible Additional Outliers: Case of Interval Uncertainty. Reliable Computing. 2021 Jun;28:100-106.
Dbouk, Hani ; Schön, Steffen ; Neumann, Ingo et al. / How to Detect Possible Additional Outliers : Case of Interval Uncertainty. In: Reliable Computing. 2021 ; Vol. 28. pp. 100-106.
Download
@article{7506ca8889fd4c7b8882bc94e6e3890b,
title = "How to Detect Possible Additional Outliers: Case of Interval Uncertainty",
abstract = "In many practical situations, measurements are characterized by interval uncertainty - namely, based on each measurement result, the only information that we have about the actual value of the measured quantity is that this value belongs to some interval. If several such intervals - corresponding to measuring the same quantity - have an empty intersection, this means that at least one of the corresponding measurement results is an outlier, caused by a malfunction of the measuring instrument. From the purely mathematical viewpoint, if the intersection is non-empty, there is no reason to be suspicious. However, from the practical viewpoint, if the intersection is too narrow - i.e., almost empty - then we should also be suspicious, and mark this as an possible additional outlier case. In this paper, we describe a natural way to formalize this idea, and an algorithm for detecting such additional possible outliers.",
keywords = "interval uncertainty, outliers, probabilistic approach",
author = "Hani Dbouk and Steffen Sch{\"o}n and Ingo Neumann and Vladik Kreinovichy",
note = "Funding information: This work was supported by the German Research Foundation (DFG) as a part of the Research Training Group i.c.sens (grant GRK2159) and by the Institute of Geodesy of the Leibniz University of Hannover. It was also supported in part by the US National Science Foundation grants 1623190 (A Model of Change for Preparing a New Generation for Professional Practice in Computer Science) and HRD-1242122 (Cyber-ShARE Center of Excellence). This paper was written when V. Kreinovich was visiting Leibniz University of Hannover. The authors are grateful to the anonymous referees for valuable suggestions.",
year = "2021",
month = jun,
language = "English",
volume = "28",
pages = "100--106",

}

Download

TY - JOUR

T1 - How to Detect Possible Additional Outliers

T2 - Case of Interval Uncertainty

AU - Dbouk, Hani

AU - Schön, Steffen

AU - Neumann, Ingo

AU - Kreinovichy, Vladik

N1 - Funding information: This work was supported by the German Research Foundation (DFG) as a part of the Research Training Group i.c.sens (grant GRK2159) and by the Institute of Geodesy of the Leibniz University of Hannover. It was also supported in part by the US National Science Foundation grants 1623190 (A Model of Change for Preparing a New Generation for Professional Practice in Computer Science) and HRD-1242122 (Cyber-ShARE Center of Excellence). This paper was written when V. Kreinovich was visiting Leibniz University of Hannover. The authors are grateful to the anonymous referees for valuable suggestions.

PY - 2021/6

Y1 - 2021/6

N2 - In many practical situations, measurements are characterized by interval uncertainty - namely, based on each measurement result, the only information that we have about the actual value of the measured quantity is that this value belongs to some interval. If several such intervals - corresponding to measuring the same quantity - have an empty intersection, this means that at least one of the corresponding measurement results is an outlier, caused by a malfunction of the measuring instrument. From the purely mathematical viewpoint, if the intersection is non-empty, there is no reason to be suspicious. However, from the practical viewpoint, if the intersection is too narrow - i.e., almost empty - then we should also be suspicious, and mark this as an possible additional outlier case. In this paper, we describe a natural way to formalize this idea, and an algorithm for detecting such additional possible outliers.

AB - In many practical situations, measurements are characterized by interval uncertainty - namely, based on each measurement result, the only information that we have about the actual value of the measured quantity is that this value belongs to some interval. If several such intervals - corresponding to measuring the same quantity - have an empty intersection, this means that at least one of the corresponding measurement results is an outlier, caused by a malfunction of the measuring instrument. From the purely mathematical viewpoint, if the intersection is non-empty, there is no reason to be suspicious. However, from the practical viewpoint, if the intersection is too narrow - i.e., almost empty - then we should also be suspicious, and mark this as an possible additional outlier case. In this paper, we describe a natural way to formalize this idea, and an algorithm for detecting such additional possible outliers.

KW - interval uncertainty

KW - outliers

KW - probabilistic approach

UR - http://www.scopus.com/inward/record.url?scp=85174703928&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:85174703928

VL - 28

SP - 100

EP - 106

JO - Reliable Computing

JF - Reliable Computing

SN - 1385-3139

ER -

By the same author(s)