Intrinsic indicators for numerical data quality

Milen S. Marev; Ernesto Compatangelo; Wamberto W. Vasconcelos

doi:10.5220/0009411403410348

Intrinsic indicators for numerical data quality

Milen S. Marev
, Ernesto Compatangelo
, Wamberto W. Vasconcelos

Computing Science

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

2 Citations (Scopus)

Abstract

This paper focuses on data quality indicators conceived to measure the quality of numerical datasets. We have devised a set of three different indicators, namely Intrinsic Quality, Distance-based Quality Factor and Information Entropy. The results of quality measures based on these indicators can be used in further data processing, helping to support actual data quality improvements. We argue that the proposed indicators can adequately capture in a quantitative way the impact of different numerical data quality issues including (but not limited to) gaps, noise or outliers.

Original language	English
Title of host publication	IoTBDS 2020 - Proceedings of the 5th International Conference on Internet of Things, Big Data and Security
Editors	Gary Wills, Peter Kacsuk, Victor Chang
Publisher	SciTePress
Pages	341-348
Number of pages	8
ISBN (Electronic)	9789897584268
DOIs	https://doi.org/10.5220/0009411403410348
Publication status	Published - 2020
Event	5th International Conference on Internet of Things, Big Data and Security, IoTBDS 2020 - Virtual, Online Duration: 7 May 2020 → 9 May 2020

Conference

Conference	5th International Conference on Internet of Things, Big Data and Security, IoTBDS 2020
City	Virtual, Online
Period	7/05/20 → 9/05/20

Funding

This research is funded by EPSRC Doctoral Training Partnership 2016-2017 University of Aberdeen with award number: EP/N509814/1

Keywords

Data Quality
Data Quality Indicators
Intrinsic Data Quality
Numerical Data Quality
Pre-processing

Access to Document

10.5220/0009411403410348

Cite this

Intrinsic indicators for numerical data quality. / Marev, Milen S.; Compatangelo, Ernesto; Vasconcelos, Wamberto W.
IoTBDS 2020 - Proceedings of the 5th International Conference on Internet of Things, Big Data and Security. ed. / Gary Wills; Peter Kacsuk; Victor Chang. SciTePress, 2020. p. 341-348.

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

Marev, MS, Compatangelo, E & Vasconcelos, WW 2020, Intrinsic indicators for numerical data quality. in G Wills, P Kacsuk & V Chang (eds), IoTBDS 2020 - Proceedings of the 5th International Conference on Internet of Things, Big Data and Security. SciTePress, pp. 341-348, 5th International Conference on Internet of Things, Big Data and Security, IoTBDS 2020, Virtual, Online, 7/05/20. https://doi.org/10.5220/0009411403410348

@inproceedings{a97b8b89524f4b2b98e53c4518ed7f7a,

title = "Intrinsic indicators for numerical data quality",

abstract = "This paper focuses on data quality indicators conceived to measure the quality of numerical datasets. We have devised a set of three different indicators, namely Intrinsic Quality, Distance-based Quality Factor and Information Entropy. The results of quality measures based on these indicators can be used in further data processing, helping to support actual data quality improvements. We argue that the proposed indicators can adequately capture in a quantitative way the impact of different numerical data quality issues including (but not limited to) gaps, noise or outliers.",

keywords = "Data Quality, Data Quality Indicators, Intrinsic Data Quality, Numerical Data Quality, Pre-processing",

author = "Marev, \{Milen S.\} and Ernesto Compatangelo and Vasconcelos, \{Wamberto W.\}",

year = "2020",

doi = "10.5220/0009411403410348",

language = "English",

pages = "341--348",

editor = "Gary Wills and Peter Kacsuk and Victor Chang",

booktitle = "IoTBDS 2020 - Proceedings of the 5th International Conference on Internet of Things, Big Data and Security",

publisher = "SciTePress",

note = "5th International Conference on Internet of Things, Big Data and Security, IoTBDS 2020 ; Conference date: 07-05-2020 Through 09-05-2020",

}

TY - GEN

T1 - Intrinsic indicators for numerical data quality

AU - Marev, Milen S.

AU - Compatangelo, Ernesto

AU - Vasconcelos, Wamberto W.

PY - 2020

Y1 - 2020

N2 - This paper focuses on data quality indicators conceived to measure the quality of numerical datasets. We have devised a set of three different indicators, namely Intrinsic Quality, Distance-based Quality Factor and Information Entropy. The results of quality measures based on these indicators can be used in further data processing, helping to support actual data quality improvements. We argue that the proposed indicators can adequately capture in a quantitative way the impact of different numerical data quality issues including (but not limited to) gaps, noise or outliers.

AB - This paper focuses on data quality indicators conceived to measure the quality of numerical datasets. We have devised a set of three different indicators, namely Intrinsic Quality, Distance-based Quality Factor and Information Entropy. The results of quality measures based on these indicators can be used in further data processing, helping to support actual data quality improvements. We argue that the proposed indicators can adequately capture in a quantitative way the impact of different numerical data quality issues including (but not limited to) gaps, noise or outliers.

KW - Data Quality

KW - Data Quality Indicators

KW - Intrinsic Data Quality

KW - Numerical Data Quality

KW - Pre-processing

UR - https://www.scopus.com/pages/publications/85089485024

U2 - 10.5220/0009411403410348

DO - 10.5220/0009411403410348

M3 - Published conference contribution

AN - SCOPUS:85089485024

SP - 341

EP - 348

BT - IoTBDS 2020 - Proceedings of the 5th International Conference on Internet of Things, Big Data and Security

A2 - Wills, Gary

A2 - Kacsuk, Peter

A2 - Chang, Victor

PB - SciTePress

T2 - 5th International Conference on Internet of Things, Big Data and Security, IoTBDS 2020

Y2 - 7 May 2020 through 9 May 2020

ER -

Intrinsic indicators for numerical data quality

Abstract

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this