SemTex: A Hybrid Approach for Semantic Table Interpretation

Emil G. Henriksen; Alan M. Khorsid; Esben Nielsen; Adam M. Stück; Andreas S. Sørensen; Olivier Pelgrin

SemTex: A Hybrid Approach for Semantic Table Interpretation

Emil G. Henriksen, Alan M. Khorsid, Esben Nielsen, Adam M. Stück, Andreas S. Sørensen, Olivier Pelgrin

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

8 Downloads (Pure)

Abstract

Accurate Semantic Table Interpretation (STI) annotation has a significant impact on interpreting unlabeled data for data analysis. In this paper, we present SemTex, a system for solving the three tasks Cell Entity Annotation (CEA), Cell Type Annotation (CTA) and Cell Property Annotation (CPA) in the Semantic Web Challenge on Tabular Data to Knowledge Graph Matching (SemTab). We utilise a hybrid approach combining analysis of relationships in knowledge graphs and gradient boosting for annotation. We document and benchmark our performance using datasets from the SemTab challenge 2022 and 2023. Our approach yields competitive results compared to the current state-of-the-art tools in all three tasks.

Original language	English
Journal	CEUR Workshop Proceedings
Volume	3557
Pages (from-to)	38-49
Number of pages	12
ISSN	1613-0073
Publication status	Published - 2023
Event	2023 Semantic Web Challenge on Tabular Data to Knowledge Graph Matching, SemTab 2023 - Athens, Greece Duration: 6 Nov 2023 → 10 Nov 2023

Conference

Conference	2023 Semantic Web Challenge on Tabular Data to Knowledge Graph Matching, SemTab 2023
Country/Territory	Greece
City	Athens
Period	06/11/2023 → 10/11/2023

Bibliographical note

Publisher Copyright:
© 2023 CEUR-WS. All rights reserved.

Keywords

Gradient Boosting
Knowledge Graph
Semantic Table Interpretation
SemTab Challenge
Tabular Data

Access to Document

Open Access articleFinal published version, 489 KBLicence: CC BY 4.0

https://ceur-ws.org/Vol-3557/paper3.pdfLicence: CC BY 4.0

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

@inproceedings{075d9dc2f370433d8dcbeb9646ad0467,

title = "SemTex: A Hybrid Approach for Semantic Table Interpretation",

abstract = "Accurate Semantic Table Interpretation (STI) annotation has a significant impact on interpreting unlabeled data for data analysis. In this paper, we present SemTex, a system for solving the three tasks Cell Entity Annotation (CEA), Cell Type Annotation (CTA) and Cell Property Annotation (CPA) in the Semantic Web Challenge on Tabular Data to Knowledge Graph Matching (SemTab). We utilise a hybrid approach combining analysis of relationships in knowledge graphs and gradient boosting for annotation. We document and benchmark our performance using datasets from the SemTab challenge 2022 and 2023. Our approach yields competitive results compared to the current state-of-the-art tools in all three tasks.",

keywords = "Gradient Boosting, Knowledge Graph, Semantic Table Interpretation, SemTab Challenge, Tabular Data",

author = "Henriksen, {Emil G.} and Khorsid, {Alan M.} and Esben Nielsen and St{\"u}ck, {Adam M.} and S{\o}rensen, {Andreas S.} and Olivier Pelgrin",

note = "Publisher Copyright: {\textcopyright} 2023 CEUR-WS. All rights reserved.; 2023 Semantic Web Challenge on Tabular Data to Knowledge Graph Matching, SemTab 2023 ; Conference date: 06-11-2023 Through 10-11-2023",

year = "2023",

language = "English",

volume = "3557",

pages = "38--49",

journal = "CEUR Workshop Proceedings",

issn = "1613-0073",

publisher = "CEUR Workshop Proceedings",

}

TY - GEN

T1 - SemTex

T2 - 2023 Semantic Web Challenge on Tabular Data to Knowledge Graph Matching, SemTab 2023

AU - Henriksen, Emil G.

AU - Khorsid, Alan M.

AU - Nielsen, Esben

AU - Stück, Adam M.

AU - Sørensen, Andreas S.

AU - Pelgrin, Olivier

PY - 2023

Y1 - 2023

N2 - Accurate Semantic Table Interpretation (STI) annotation has a significant impact on interpreting unlabeled data for data analysis. In this paper, we present SemTex, a system for solving the three tasks Cell Entity Annotation (CEA), Cell Type Annotation (CTA) and Cell Property Annotation (CPA) in the Semantic Web Challenge on Tabular Data to Knowledge Graph Matching (SemTab). We utilise a hybrid approach combining analysis of relationships in knowledge graphs and gradient boosting for annotation. We document and benchmark our performance using datasets from the SemTab challenge 2022 and 2023. Our approach yields competitive results compared to the current state-of-the-art tools in all three tasks.

AB - Accurate Semantic Table Interpretation (STI) annotation has a significant impact on interpreting unlabeled data for data analysis. In this paper, we present SemTex, a system for solving the three tasks Cell Entity Annotation (CEA), Cell Type Annotation (CTA) and Cell Property Annotation (CPA) in the Semantic Web Challenge on Tabular Data to Knowledge Graph Matching (SemTab). We utilise a hybrid approach combining analysis of relationships in knowledge graphs and gradient boosting for annotation. We document and benchmark our performance using datasets from the SemTab challenge 2022 and 2023. Our approach yields competitive results compared to the current state-of-the-art tools in all three tasks.

KW - Gradient Boosting

KW - Knowledge Graph

KW - Semantic Table Interpretation

KW - SemTab Challenge

KW - Tabular Data

UR - http://www.scopus.com/inward/record.url?scp=85179134840&partnerID=8YFLogxK

M3 - Conference article in Journal

AN - SCOPUS:85179134840

SN - 1613-0073

VL - 3557

SP - 38

EP - 49

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

Y2 - 6 November 2023 through 10 November 2023

ER -

SemTex: A Hybrid Approach for Semantic Table Interpretation

Abstract

Conference

Bibliographical note

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Cite this