Details
| Original language | English |
|---|---|
| Pages (from-to) | 100-106 |
| Number of pages | 7 |
| Journal | Reliable Computing |
| Volume | 28 |
| Publication status | Published - 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
- Computer Science(all)
- Software
- Mathematics(all)
- Computational Mathematics
- Mathematics(all)
- Applied Mathematics
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Reliable Computing, Vol. 28, 06.2021, p. 100-106.
Research output: Contribution to journal › Article › Research › peer review
}
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 -