Learning sequence patterns in knowledge graph triples to predict inconsistencies

Mahmoud Elbattah; Conor Ryan

Learning sequence patterns in knowledge graph triples to predict inconsistencies

Mahmoud Elbattah, Conor Ryan

Department of Computer Science & Information Systems

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

Abstract

The current trend towards the Semantic Web and Linked Data has resulted in an unprecedented volume of data being continuously published on the Linked Open Data (LOD) cloud. Massive Knowledge Graphs (KGs) are increasingly constructed and enriched based on large amounts of unstructured data. However, the data quality of KGs can still suffer from a variety of inconsistencies, misinterpretations or incomplete information as well. This study investigates the feasibility of utilising the subject-predicate-object (SPO) structure of KG triples to detect possible inconsistencies. The key idea is hinged on using the Freebase-defined entity types for extracting the unique SPO patterns in the KG. Using Machine learning, the problem of predicting inconsistencies could be approached as a sequence classification task. The approach applicability was experimented using a subset of the Freebase KG, which included about 6M triples. The experiments proved promising results using Convnet and LSTM models for detecting inconsistent sequences.

Original language	English
Title of host publication	IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management
Editors	Jorge Bernardino, Ana Salgado, Joaquim Filipe
Publisher	SciTePress
Pages	435-441
Number of pages	7
ISBN (Electronic)	9789897583827
Publication status	Published - 2019
Event	11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management, IC3K 2019 - Vienna, Austria Duration: 17 Sep 2019 → 19 Sep 2019

Publication series

Name	IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management
Volume	3

Conference

Conference	11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management, IC3K 2019
Country/Territory	Austria
City	Vienna
Period	17/09/19 → 19/09/19

Keywords

Knowledge Graphs
Knowledgebase
Machine Learning
Semantic Web

Cite this

Elbattah, M., & Ryan, C. (2019). Learning sequence patterns in knowledge graph triples to predict inconsistencies. In J. Bernardino, A. Salgado, & J. Filipe (Eds.), IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (pp. 435-441). (IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management; Vol. 3). SciTePress.

Elbattah, Mahmoud ; Ryan, Conor. / Learning sequence patterns in knowledge graph triples to predict inconsistencies. IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management. editor / Jorge Bernardino ; Ana Salgado ; Joaquim Filipe. SciTePress, 2019. pp. 435-441 (IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management).

@inproceedings{eed1aaec8bc84d2c89c67114945a42b3,

title = "Learning sequence patterns in knowledge graph triples to predict inconsistencies",

abstract = "The current trend towards the Semantic Web and Linked Data has resulted in an unprecedented volume of data being continuously published on the Linked Open Data (LOD) cloud. Massive Knowledge Graphs (KGs) are increasingly constructed and enriched based on large amounts of unstructured data. However, the data quality of KGs can still suffer from a variety of inconsistencies, misinterpretations or incomplete information as well. This study investigates the feasibility of utilising the subject-predicate-object (SPO) structure of KG triples to detect possible inconsistencies. The key idea is hinged on using the Freebase-defined entity types for extracting the unique SPO patterns in the KG. Using Machine learning, the problem of predicting inconsistencies could be approached as a sequence classification task. The approach applicability was experimented using a subset of the Freebase KG, which included about 6M triples. The experiments proved promising results using Convnet and LSTM models for detecting inconsistent sequences.",

keywords = "Knowledge Graphs, Knowledgebase, Machine Learning, Semantic Web",

author = "Mahmoud Elbattah and Conor Ryan",

note = "Publisher Copyright: Copyright {\textcopyright} 2019 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved; 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management, IC3K 2019 ; Conference date: 17-09-2019 Through 19-09-2019",

year = "2019",

language = "English",

series = "IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management",

publisher = "SciTePress",

pages = "435--441",

editor = "Jorge Bernardino and Ana Salgado and Joaquim Filipe",

booktitle = "IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management",

}

Elbattah, M & Ryan, C 2019, Learning sequence patterns in knowledge graph triples to predict inconsistencies. in J Bernardino, A Salgado & J Filipe (eds), IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management. IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management, vol. 3, SciTePress, pp. 435-441, 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management, IC3K 2019, Vienna, Austria, 17/09/19.

Learning sequence patterns in knowledge graph triples to predict inconsistencies. / Elbattah, Mahmoud; Ryan, Conor.
IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management. ed. / Jorge Bernardino; Ana Salgado; Joaquim Filipe. SciTePress, 2019. p. 435-441 (IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management; Vol. 3).

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

TY - GEN

T1 - Learning sequence patterns in knowledge graph triples to predict inconsistencies

AU - Elbattah, Mahmoud

AU - Ryan, Conor

PY - 2019

Y1 - 2019

N2 - The current trend towards the Semantic Web and Linked Data has resulted in an unprecedented volume of data being continuously published on the Linked Open Data (LOD) cloud. Massive Knowledge Graphs (KGs) are increasingly constructed and enriched based on large amounts of unstructured data. However, the data quality of KGs can still suffer from a variety of inconsistencies, misinterpretations or incomplete information as well. This study investigates the feasibility of utilising the subject-predicate-object (SPO) structure of KG triples to detect possible inconsistencies. The key idea is hinged on using the Freebase-defined entity types for extracting the unique SPO patterns in the KG. Using Machine learning, the problem of predicting inconsistencies could be approached as a sequence classification task. The approach applicability was experimented using a subset of the Freebase KG, which included about 6M triples. The experiments proved promising results using Convnet and LSTM models for detecting inconsistent sequences.

AB - The current trend towards the Semantic Web and Linked Data has resulted in an unprecedented volume of data being continuously published on the Linked Open Data (LOD) cloud. Massive Knowledge Graphs (KGs) are increasingly constructed and enriched based on large amounts of unstructured data. However, the data quality of KGs can still suffer from a variety of inconsistencies, misinterpretations or incomplete information as well. This study investigates the feasibility of utilising the subject-predicate-object (SPO) structure of KG triples to detect possible inconsistencies. The key idea is hinged on using the Freebase-defined entity types for extracting the unique SPO patterns in the KG. Using Machine learning, the problem of predicting inconsistencies could be approached as a sequence classification task. The approach applicability was experimented using a subset of the Freebase KG, which included about 6M triples. The experiments proved promising results using Convnet and LSTM models for detecting inconsistent sequences.

KW - Knowledge Graphs

KW - Knowledgebase

KW - Machine Learning

KW - Semantic Web

UR - http://www.scopus.com/inward/record.url?scp=85074148286&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85074148286

T3 - IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management

SP - 435

EP - 441

BT - IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management

A2 - Bernardino, Jorge

A2 - Salgado, Ana

A2 - Filipe, Joaquim

PB - SciTePress

T2 - 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management, IC3K 2019

Y2 - 17 September 2019 through 19 September 2019

ER -

Elbattah M, Ryan C. Learning sequence patterns in knowledge graph triples to predict inconsistencies. In Bernardino J, Salgado A, Filipe J, editors, IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management. SciTePress. 2019. p. 435-441. (IC3K 2019 - Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management).

Learning sequence patterns in knowledge graph triples to predict inconsistencies

Abstract

Publication series

Conference

Keywords

Other files and links

Fingerprint

Cite this