Do you catch my drift? On the usage of embedding methods to measure concept shift in knowledge graphs

Stella Verkijk; Ritten Roothaert; Romana Pernisch; Stefan Schlobach

doi:10.1145/3587259.3627555

Do you catch my drift? On the usage of embedding methods to measure concept shift in knowledge graphs

Stella Verkijk, Ritten Roothaert, Romana Pernisch, Stefan Schlobach

Research output: Chapter in Book / Report / Conference proceeding › Conference contribution › Academic › peer-review

Abstract

Automatically detecting and measuring differences between evolving Knowledge Graphs (KGs) has been a topic of investigation for years. With the rising popularity of embedding methods, we investigate the possibility of using embeddings to detect Concept Shift in evolving KGs. Specifically, we go deeper into the usage of nearest neighbour set comparison as the basis for a similarity measure, and show why this approach is conceptually problematic. As an alternative, we explore the possibility of using clustering methods. This paper serves to (i) inform the community about the challenges that arise when using KG embeddings for the comparison of different versions of a KG specifically, (ii) investigate how this is supported by theories on knowledge representation and semantic representation in NLP and (iii) take the first steps into the direction of valuable representation of semantics within KGs for comparison.

Original language	English
Title of host publication	K-CAP ’23
Subtitle of host publication	Proceedings of the 12th Knowledge Capture Conference 2023
Publisher	Association for Computing Machinery, Inc
Pages	70-74
Number of pages	5
ISBN (Electronic)	9798400701412
ISBN (Print)	979840070141
DOIs	https://doi.org/10.1145/3587259.3627555
Publication status	Published - Dec 2023
Event	12th ACM International Conference on Knowledge Capture, K-CAP 2023 - Pensacola, United States Duration: 5 Dec 2023 → 7 Dec 2023

Conference

Conference	12th ACM International Conference on Knowledge Capture, K-CAP 2023
Country/Territory	United States
City	Pensacola
Period	5/12/23 → 7/12/23

Bibliographical note

Publisher Copyright:
© 2023 Owner/Author.

Keywords

Concept Shift
Knowledge Graph Embeddings
NLP
Semantics

Access to Document

10.1145/3587259.3627555

Persistent URL (handle)

Persistent link

Cite this

@inproceedings{250f528453574a9dbc1bbc4772cb38cc,

title = "Do you catch my drift? On the usage of embedding methods to measure concept shift in knowledge graphs",

abstract = "Automatically detecting and measuring differences between evolving Knowledge Graphs (KGs) has been a topic of investigation for years. With the rising popularity of embedding methods, we investigate the possibility of using embeddings to detect Concept Shift in evolving KGs. Specifically, we go deeper into the usage of nearest neighbour set comparison as the basis for a similarity measure, and show why this approach is conceptually problematic. As an alternative, we explore the possibility of using clustering methods. This paper serves to (i) inform the community about the challenges that arise when using KG embeddings for the comparison of different versions of a KG specifically, (ii) investigate how this is supported by theories on knowledge representation and semantic representation in NLP and (iii) take the first steps into the direction of valuable representation of semantics within KGs for comparison.",

keywords = "Concept Shift, Knowledge Graph Embeddings, NLP, Semantics",

author = "Stella Verkijk and Ritten Roothaert and Romana Pernisch and Stefan Schlobach",

note = "Publisher Copyright: {\textcopyright} 2023 Owner/Author.; 12th ACM International Conference on Knowledge Capture, K-CAP 2023 ; Conference date: 05-12-2023 Through 07-12-2023",

year = "2023",

month = dec,

doi = "10.1145/3587259.3627555",

language = "English",

isbn = "979840070141",

pages = "70--74",

booktitle = "K-CAP {\textquoteright}23",

publisher = "Association for Computing Machinery, Inc",

}

Verkijk, S, Roothaert, R , Pernisch, R & Schlobach, S 2023, Do you catch my drift? On the usage of embedding methods to measure concept shift in knowledge graphs. in K-CAP ’23: Proceedings of the 12th Knowledge Capture Conference 2023. Association for Computing Machinery, Inc, pp. 70-74, 12th ACM International Conference on Knowledge Capture, K-CAP 2023, Pensacola, United States, 5/12/23. https://doi.org/10.1145/3587259.3627555

Do you catch my drift? On the usage of embedding methods to measure concept shift in knowledge graphs. / Verkijk, Stella; Roothaert, Ritten ; Pernisch, Romana et al.
K-CAP ’23: Proceedings of the 12th Knowledge Capture Conference 2023. Association for Computing Machinery, Inc, 2023. p. 70-74.

Research output: Chapter in Book / Report / Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Do you catch my drift? On the usage of embedding methods to measure concept shift in knowledge graphs

AU - Verkijk, Stella

AU - Roothaert, Ritten

AU - Pernisch, Romana

AU - Schlobach, Stefan

PY - 2023/12

Y1 - 2023/12

N2 - Automatically detecting and measuring differences between evolving Knowledge Graphs (KGs) has been a topic of investigation for years. With the rising popularity of embedding methods, we investigate the possibility of using embeddings to detect Concept Shift in evolving KGs. Specifically, we go deeper into the usage of nearest neighbour set comparison as the basis for a similarity measure, and show why this approach is conceptually problematic. As an alternative, we explore the possibility of using clustering methods. This paper serves to (i) inform the community about the challenges that arise when using KG embeddings for the comparison of different versions of a KG specifically, (ii) investigate how this is supported by theories on knowledge representation and semantic representation in NLP and (iii) take the first steps into the direction of valuable representation of semantics within KGs for comparison.

AB - Automatically detecting and measuring differences between evolving Knowledge Graphs (KGs) has been a topic of investigation for years. With the rising popularity of embedding methods, we investigate the possibility of using embeddings to detect Concept Shift in evolving KGs. Specifically, we go deeper into the usage of nearest neighbour set comparison as the basis for a similarity measure, and show why this approach is conceptually problematic. As an alternative, we explore the possibility of using clustering methods. This paper serves to (i) inform the community about the challenges that arise when using KG embeddings for the comparison of different versions of a KG specifically, (ii) investigate how this is supported by theories on knowledge representation and semantic representation in NLP and (iii) take the first steps into the direction of valuable representation of semantics within KGs for comparison.

KW - Concept Shift

KW - Knowledge Graph Embeddings

KW - NLP

KW - Semantics

UR - http://www.scopus.com/inward/record.url?scp=85180365207&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85180365207&partnerID=8YFLogxK

U2 - 10.1145/3587259.3627555

DO - 10.1145/3587259.3627555

M3 - Conference contribution

AN - SCOPUS:85180365207

SN - 979840070141

SP - 70

EP - 74

BT - K-CAP ’23

PB - Association for Computing Machinery, Inc

T2 - 12th ACM International Conference on Knowledge Capture, K-CAP 2023

Y2 - 5 December 2023 through 7 December 2023

ER -

Do you catch my drift? On the usage of embedding methods to measure concept shift in knowledge graphs

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Persistent URL (handle)

Other files and links

Fingerprint

Cite this