Provenance Network Analytics: An approach to data analytics using data provenance

Trung Dong Huynh; Mark Ebden; Joel Fischer; Stephen Roberts; Luc Moreau

doi:10.1007/s10618-017-0549-3

Provenance Network Analytics: An approach to data analytics using data provenance

Trung Dong Huynh, Mark Ebden, Joel Fischer, Stephen Roberts, Luc Moreau

Research output: Contribution to journal › Article › peer-review

15 Citations (Scopus)

283 Downloads (Pure)

Abstract

Provenance network analytics is a novel data analytics approach that helps infer properties of data, such as quality or importance, from their provenance. Instead of analysing application data, which are typically domain-dependent, it analyses the data's provenance as represented using the World Wide Web Consortium's domain-agnostic PROV data model. Specifically, the approach proposes a number of network metrics for provenance data and applies established machine learning techniques over such metrics to build predictive models for some key properties of data. Applying this method to the provenance of real-world data from three different applications, we show that it can successfully identify the owners of provenance documents, assess the quality of crowdsourced data, and identify instructions from chat messages in an alternate-reality game with high levels of accuracy. By so doing, we demonstrate the different ways the proposed provenance network metrics can be used in analysing data, providing the foundation for provenance-based data analytics.

Original language	English
Pages (from-to)	708-735
Journal	DATA MINING AND KNOWLEDGE DISCOVERY
Volume	32
Early online date	15 Feb 2018
DOIs	https://doi.org/10.1007/s10618-017-0549-3
Publication status	Published - 30 May 2018

Keywords

data provenance
data analytics
network metrics
graph classification

Access to Document

10.1007/s10618-017-0549-3Licence: CC BY

Provenance Network Analytics_HUYNH_Firstonline15February2018_GOLD CP (CC BY)Final published version, 1.12 MBLicence: CC BY

Cite this

@article{a3954c85d9ae4275aa59e317413b77b9,

title = "Provenance Network Analytics: An approach to data analytics using data provenance",

abstract = "Provenance network analytics is a novel data analytics approach that helps infer properties of data, such as quality or importance, from their provenance. Instead of analysing application data, which are typically domain-dependent, it analyses the data's provenance as represented using the World Wide Web Consortium's domain-agnostic PROV data model. Specifically, the approach proposes a number of network metrics for provenance data and applies established machine learning techniques over such metrics to build predictive models for some key properties of data. Applying this method to the provenance of real-world data from three different applications, we show that it can successfully identify the owners of provenance documents, assess the quality of crowdsourced data, and identify instructions from chat messages in an alternate-reality game with high levels of accuracy. By so doing, we demonstrate the different ways the proposed provenance network metrics can be used in analysing data, providing the foundation for provenance-based data analytics.",

keywords = "data provenance, data analytics, network metrics, graph classification",

author = "Huynh, {Trung Dong} and Mark Ebden and Joel Fischer and Stephen Roberts and Luc Moreau",

year = "2018",

month = may,

day = "30",

doi = "10.1007/s10618-017-0549-3",

language = "English",

volume = "32",

pages = "708--735",

journal = "DATA MINING AND KNOWLEDGE DISCOVERY",

issn = "1384-5810",

publisher = "Springer Netherlands",

}

TY - JOUR

T1 - Provenance Network Analytics

T2 - An approach to data analytics using data provenance

AU - Huynh, Trung Dong

AU - Ebden, Mark

AU - Fischer, Joel

AU - Roberts, Stephen

AU - Moreau, Luc

PY - 2018/5/30

Y1 - 2018/5/30

N2 - Provenance network analytics is a novel data analytics approach that helps infer properties of data, such as quality or importance, from their provenance. Instead of analysing application data, which are typically domain-dependent, it analyses the data's provenance as represented using the World Wide Web Consortium's domain-agnostic PROV data model. Specifically, the approach proposes a number of network metrics for provenance data and applies established machine learning techniques over such metrics to build predictive models for some key properties of data. Applying this method to the provenance of real-world data from three different applications, we show that it can successfully identify the owners of provenance documents, assess the quality of crowdsourced data, and identify instructions from chat messages in an alternate-reality game with high levels of accuracy. By so doing, we demonstrate the different ways the proposed provenance network metrics can be used in analysing data, providing the foundation for provenance-based data analytics.

AB - Provenance network analytics is a novel data analytics approach that helps infer properties of data, such as quality or importance, from their provenance. Instead of analysing application data, which are typically domain-dependent, it analyses the data's provenance as represented using the World Wide Web Consortium's domain-agnostic PROV data model. Specifically, the approach proposes a number of network metrics for provenance data and applies established machine learning techniques over such metrics to build predictive models for some key properties of data. Applying this method to the provenance of real-world data from three different applications, we show that it can successfully identify the owners of provenance documents, assess the quality of crowdsourced data, and identify instructions from chat messages in an alternate-reality game with high levels of accuracy. By so doing, we demonstrate the different ways the proposed provenance network metrics can be used in analysing data, providing the foundation for provenance-based data analytics.

KW - data provenance

KW - data analytics

KW - network metrics

KW - graph classification

UR - http://www.scopus.com/inward/record.url?scp=85042114156&partnerID=8YFLogxK

U2 - 10.1007/s10618-017-0549-3

DO - 10.1007/s10618-017-0549-3

M3 - Article

SN - 1384-5810

VL - 32

SP - 708

EP - 735

JO - DATA MINING AND KNOWLEDGE DISCOVERY

JF - DATA MINING AND KNOWLEDGE DISCOVERY

ER -

Provenance Network Analytics: An approach to data analytics using data provenance

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this