From why-provenance to why+provenance:  Towards addressing deep data explanations in Data-Centric AI

Paolo Missier; Riccardo Torlone

From why-provenance to why+provenance: Towards addressing deep data explanations in Data-Centric AI

Paolo Missier^*
, Riccardo Torlone

^*Corresponding author for this work

Engineering and Physical Sciences

Research output: Contribution to journal › Conference article › peer-review

87 Downloads (Pure)

Abstract

In this position paper we discuss the problem of exploiting data provenance to provide explanations in data-centric AI processes, where the emphasis of model development is placed on the quality of data. In particular, we show how a classification of the main operators used in the data preparation phase provides an effective and powerful means for the production of increasingly detailed explanations at the needed level of data granularity.

Original language	English
Pages (from-to)	508-517
Number of pages	10
Journal	CEUR Workshop Proceedings
Volume	3741
Publication status	Published - 13 Aug 2024
Event	32nd Symposium on Advanced Database Systems 2024 - Villasimius, Italy Duration: 23 Jun 2024 → 26 Jun 2024 https://sebd2024.unica.it/

Access to Document

MissierP2024Why-provenanceFinal published version, 1.19 MBLicence: Creative Commons: Attribution (CC BY)

https://ceur-ws.org/Vol-3741/paper11.pdfLicence: Creative Commons: Attribution (CC BY)

Cite this

@article{8e2d1fe9cfdc4e7790e661650b40c416,

title = "From why-provenance to why+provenance: Towards addressing deep data explanations in Data-Centric AI",

abstract = "In this position paper we discuss the problem of exploiting data provenance to provide explanations in data-centric AI processes, where the emphasis of model development is placed on the quality of data. In particular, we show how a classification of the main operators used in the data preparation phase provides an effective and powerful means for the production of increasingly detailed explanations at the needed level of data granularity.",

author = "Paolo Missier and Riccardo Torlone",

year = "2024",

month = aug,

day = "13",

language = "English",

volume = "3741",

pages = "508--517",

journal = "CEUR Workshop Proceedings",

issn = "1613-0073",

publisher = "RWTH Aachen University",

note = "32nd Symposium on Advanced Database Systems 2024, SEBD 2024 ; Conference date: 23-06-2024 Through 26-06-2024",

url = "https://sebd2024.unica.it/",

}

TY - JOUR

T1 - From why-provenance to why+provenance

T2 - 32nd Symposium on Advanced Database Systems 2024

AU - Missier, Paolo

AU - Torlone, Riccardo

PY - 2024/8/13

Y1 - 2024/8/13

N2 - In this position paper we discuss the problem of exploiting data provenance to provide explanations in data-centric AI processes, where the emphasis of model development is placed on the quality of data. In particular, we show how a classification of the main operators used in the data preparation phase provides an effective and powerful means for the production of increasingly detailed explanations at the needed level of data granularity.

AB - In this position paper we discuss the problem of exploiting data provenance to provide explanations in data-centric AI processes, where the emphasis of model development is placed on the quality of data. In particular, we show how a classification of the main operators used in the data preparation phase provides an effective and powerful means for the production of increasingly detailed explanations at the needed level of data granularity.

UR - https://ceur-ws.org/Vol-3741/

M3 - Conference article

SN - 1613-0073

VL - 3741

SP - 508

EP - 517

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

Y2 - 23 June 2024 through 26 June 2024

ER -

From why-provenance to why+provenance: Towards addressing deep data explanations in Data-Centric AI

Abstract

Access to Document

Fingerprint

Cite this