Language-Based Testing for Knowledge Graphs

Tobias John; Einar Broch Johnsen; Eduard Kamburjan; Dominic Steinhöfel

doi:10.1007/978-3-031-94578-6_2

Language-Based Testing for Knowledge Graphs

Tobias John
, Einar Broch Johnsen
, Eduard Kamburjan
, Dominic Steinhöfel

University of Oslo
CISPA Helmholz Center for Information Security

Research output: Journal Article or Conference Article in Journal › Conference article › Research › peer-review

Abstract

Knowledge graphs rely on a vast ecosystem of software tools, such as parsers, APIs and reasoners. Yet, tool developers have little support to ensure tool reliability. Here, we demonstrate how recent advantages in test case generation for highly structured and constrained inputs can support software developers in the semantic web field. We develop input generators for RDF Turtle and the OWL EL profile, and report on numerous bugs we found in parsers, reasoners and APIs of widely used tools and libraries, as well as imprecisions in standards and documentation. We provide actionable insights on using automated testing to increase the reliability of software tools for knowledge graphs.

Original language	English
Book series	Lecture Notes in Computer Science
Volume	15719
Pages (from-to)	24-46
Number of pages	23
DOIs	https://doi.org/10.1007/978-3-031-94578-6_2
Publication status	Published - 31 May 2025
Event	European Semantic Web Conference - Portoroz, Slovenia Duration: 1 Jun 2025 → 5 Jun 2025 Conference number: 22 https://link.springer.com/book/10.1007/978-3-031-94575-5

Conference

Conference	European Semantic Web Conference
Number	22
Country/Territory	Slovenia
City	Portoroz
Period	01/06/2025 → 05/06/2025
Internet address	https://link.springer.com/book/10.1007/978-3-031-94575-5

Access to Document

10.1007/978-3-031-94578-6_2

Cite this

@inproceedings{8aab03d19f1c477dbdbe4f6d4fe7ec99,

title = "Language-Based Testing for Knowledge Graphs",

abstract = "Knowledge graphs rely on a vast ecosystem of software tools, such as parsers, APIs and reasoners. Yet, tool developers have little support to ensure tool reliability. Here, we demonstrate how recent advantages in test case generation for highly structured and constrained inputs can support software developers in the semantic web field. We develop input generators for RDF Turtle and the OWL EL profile, and report on numerous bugs we found in parsers, reasoners and APIs of widely used tools and libraries, as well as imprecisions in standards and documentation. We provide actionable insights on using automated testing to increase the reliability of software tools for knowledge graphs.",

keywords = "Automated Testing, OWL EL, RDF, Reasoning",

author = "Tobias John and Johnsen, \{Einar Broch\} and Eduard Kamburjan and Dominic Steinh{\"o}fel",

year = "2025",

month = may,

day = "31",

doi = "10.1007/978-3-031-94578-6\_2",

language = "English",

volume = "15719",

pages = "24--46",

journal = "Lecture Notes in Computer Science",

issn = "1611-3349",

publisher = "Springer",

note = "European Semantic Web Conference, ESWC ; Conference date: 01-06-2025 Through 05-06-2025",

url = "https://link.springer.com/book/10.1007/978-3-031-94575-5",

}

TY - GEN

T1 - Language-Based Testing for Knowledge Graphs

AU - John, Tobias

AU - Johnsen, Einar Broch

AU - Kamburjan, Eduard

AU - Steinhöfel, Dominic

N1 - Conference code: 22

PY - 2025/5/31

Y1 - 2025/5/31

N2 - Knowledge graphs rely on a vast ecosystem of software tools, such as parsers, APIs and reasoners. Yet, tool developers have little support to ensure tool reliability. Here, we demonstrate how recent advantages in test case generation for highly structured and constrained inputs can support software developers in the semantic web field. We develop input generators for RDF Turtle and the OWL EL profile, and report on numerous bugs we found in parsers, reasoners and APIs of widely used tools and libraries, as well as imprecisions in standards and documentation. We provide actionable insights on using automated testing to increase the reliability of software tools for knowledge graphs.

AB - Knowledge graphs rely on a vast ecosystem of software tools, such as parsers, APIs and reasoners. Yet, tool developers have little support to ensure tool reliability. Here, we demonstrate how recent advantages in test case generation for highly structured and constrained inputs can support software developers in the semantic web field. We develop input generators for RDF Turtle and the OWL EL profile, and report on numerous bugs we found in parsers, reasoners and APIs of widely used tools and libraries, as well as imprecisions in standards and documentation. We provide actionable insights on using automated testing to increase the reliability of software tools for knowledge graphs.

KW - Automated Testing

KW - OWL EL

KW - RDF

KW - Reasoning

U2 - 10.1007/978-3-031-94578-6_2

DO - 10.1007/978-3-031-94578-6_2

M3 - Conference article

SN - 1611-3349

VL - 15719

SP - 24

EP - 46

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

T2 - European Semantic Web Conference

Y2 - 1 June 2025 through 5 June 2025

ER -

Language-Based Testing for Knowledge Graphs

Abstract

Conference

Access to Document

Fingerprint

Cite this