Paper deep dive

OntoInk: Interactive Ontology Visualization, Validation, and Reasoning

Ebrahim Norouzi, Jörg Waitelonis, Harald Sack

Year: 2026Venue: arXiv preprintArea: cs.AIType: PreprintEmbeddings: 25

Intelligence

Status: not_run | Model: - | Prompt: - | Confidence: 0%

Entities (0)

No extracted entities yet.

Relation Signals (0)

No relation signals yet.

Cypher Suggestions (0)

No Cypher suggestions yet.

Abstract

Abstract:Ontology documentation, visualization, and validation are usually carried out with separate tools. This split workflow slows down development and makes knowledge transfer harder. We present OntoInk, an open-source MkDocs plugin that brings these activities together. Within a single documentation-as-code pipeline, OntoInk renders interactive ontology diagrams, validates instance data against SHACL shapes, runs OWL\,DL reasoning, and supports inline Turtle editing. General-purpose diagram plugins for MkDocs cannot parse RDF, dereference IRIs, overlay SHACL constraints, or run OWL reasoning. Compared with standalone ontology visualization tools, OntoInk embeds interactive and editable diagrams directly into documentation pages. A live demo and source code are available at \url{this https URL}.

PDF

Open source PDF →Open local PDF →

Trouble viewing inline? Open PDF directly →

Full Text

22,415 characters extracted from source content.

Expand or collapse full text

*1 OntoInk: Interactive Ontology Visualization, Validation, and Reasoning Ebrahim Norouzi Jörg Waitelonis Harald Sack Abstract Ontology documentation, visualization, and validation are usually carried out with separate tools. This split workflow slows down development and makes knowledge transfer harder. We present OntoInk, an open-source MkDocs plugin that brings these activities together. Within a single documentation-as-code pipeline, OntoInk renders interactive ontology diagrams, validates instance data against SHACL shapes, runs OWL DL reasoning, and supports inline Turtle editing. General-purpose diagram plugins for MkDocs cannot parse RDF, dereference IRIs, overlay SHACL constraints, or run OWL reasoning. Compared with standalone ontology visualization tools, OntoInk embeds interactive and editable diagrams directly into documentation pages. A live demo and source code are available at https://ise-fizkarlsruhe.github.io/ontoink/. keywords Ontology Visualization ,Ontology Documentation ,SHACL Validation ,OWL Reasoning ,MkDocs Plugin †copyrightyear: 2026†copyright: Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (C BY 4.0).†venue: ISWC 2026 Companion Volume, October 25–29, 2026, Bari, Italy†email: ebrahim.norouzi@fiz-karlsruhe.de†email: joerg.waitelonis@fiz-karlsruhe.de†email: harald.sack@fiz-karlsruhe.de†address: FIZ Karlsruhe – Leibniz Institute for Information Infrastructure, Germany†address: Karlsruhe Institute of Technology (KIT), Institute AIFB, Germany†corresp: Corresponding author. 1 Introduction Ontologies form the formal schema layer of knowledge graphs, yet their documentation and visualization are usually carried out with different tools. A wide range of ontology visualization tools has appeared over the last decade Dudáš et al. (2018); Gomes et al. (2024); Lohmann et al. (2015); Falco et al. (2014); Chávez-Feria et al. (2022), but they produce either static diagrams or stand-alone web applications that sit apart from the documentation workflow—conceptual models drawn in one tool must be kept in sync with the OWL source by hand. HTML documentation generators such as Widoco Garijo (2017) embed a (static) WebVOWL diagram, and vocabulary-development environments such as VoCol Halilaj et al. (2016) add version control and documentation generation to CI pipelines—but their output still lacks interactive graph exploration, SHACL constraint visualization, and OWL reasoning. In parallel, the documentation-as-code idea has been widely adopted, and static site generators such as MkDocs are commonly extended with diagram plugins (D2 Terrastruct (2022), Mermaid.js Mermaid Contributors (2014)). These general-purpose approaches, however, are not ontology-aware: they cannot parse RDF, resolve IRIs to fetch labels, overlay SHACL constraints, validate instance data, or execute OWL reasoning. OntoInk11 1 OntoInk source code: https://github.com/ISE-FIZKarlsruhe/ontoink closes this gap. It treats the ontology source as the single point of truth from which visualization, validation, and reasoning are derived in a Markdown-based pipeline. RDF Turtle Beckett et al. (2014) files are embedded in Markdown through fenced code blocks. At build time, the RDF graph is parsed, SHACL Knublauch and Kontokostas (2017) constraints are extracted, and OWL DL reasoning is executed. At runtime, the page shows an interactive diagram with SHACL constraint overlays, a live RDF Turtle editor with syntax highlighting, and publication-ready export. The contributions of this work are: (1) a documentation-as-code architecture that integrates ontology visualization, SHACL validation, OWL reasoning, and inline editing in a single MkDocs pipeline; (2) a formal visual notation that overlays SHACL constraints with cardinality badges directly on the graph; and (3) a visual SHACL shape editor that authors shapes through a form interface. 2 Related Work Table 1 positions OntoInk against representative tools across three categories. Ontology visualization tools focus on rendering ontologies as diagrams or interactive visualizations. Representative examples include WebVOWL Lohmann et al. (2015), which implements the VOWL notation Lohmann et al. (2016), Graffoo Falco et al. (2014), OWL2Graffoo Duchateau and Debruyne (2026), Chowlk Chávez-Feria et al. (2022), OWLGrEd Barzdins et al. (2010), and the OntoDia Mouromtsev et al. (2015)/Reactodia Mouromtsev and others (2025) family. These tools typically render ontologies as static diagrams or stand-alone web applications that are separate from the documentation workflow. None combine SHACL overlays, OWL reasoning, and inline editing. Among them, only UnSHACLed Lieber et al. (2022) supports SHACL, but its focus is on constraint authoring rather than documentation. Documentation generators such as Widoco Garijo (2017), LODE Peroni et al. (2012), and pyLODE Car (2022) generate read-only HTML documentation from ontologies. Documentation-as-code platforms, including VoCol Halilaj et al. (2016) and OnToology Alobaid et al. (2019), extend this approach by integrating version control and continuous integration. However, they still produce static documentation pages rather than interactive documentation environments. General-purpose diagram plugins for MkDocs (D2, Mermaid) are not ontology-aware: they cannot parse RDF, dereference IRIs, overlay SHACL constraints, or run reasoning, so their diagrams drift from the source like hand-drawn figures. As shown in Table 1, OntoInk, to the best of our knowledge, is the first tool that combines interactive ontology visualization with formal SHACL constraint overlays, OWL reasoning, IRI dereferencing, and inline Turtle editing within a documentation-as-code workflow. Table 1: Feature comparison of OntoInk with related ontology visualization and documentation tools. ∙ = supported, ∘ = partial, – = not supported. Interactive Vis. SHACL Overlay OWL Reasoning Inline Editing Doc. Integration Pub.-Ready Export Protégé Musen (2015) ∘ ∘ ∙ ∙ – – WebVOWL Lohmann et al. (2015) ∙ – – – – ∘ WebVOWL Editor Wiens et al. (2018) ∙ – – ∙ – ∘ Graffoo Falco et al. (2014) – – – – – ∙ OWL2Graffoo Duchateau and Debruyne (2026) ∘ – – ∙ ∘ ∙ Chowlk Chávez-Feria et al. (2022) – – – ∘ – ∙ OWLGrEd Barzdins et al. (2010) ∘ – – ∙ – ∘ OntoDia Mouromtsev et al. (2015) ∙ – – – – – Reactodia Mouromtsev and others (2025) ∙ – – ∙ – ∘ UnSHACLed Lieber et al. (2022) ∘ ∙ – ∙ – – Widoco Garijo (2017) ∙ – – – ∙ – pyLODE Car (2022) – – – – ∙ – VoCol Halilaj et al. (2016) – – – – ∙ – OnToology Alobaid et al. (2019) ∘ – – – ∙ – OntoInk ∙ ∙ ∙ ∙ ∙ ∙ 3 System Overview OntoInk is implemented as an MkDocs plugin with a two-phase architecture: a build-time processing pipeline in Python and a runtime interactive layer in JavaScript. 3.1 Build-Time Processing The plugin is triggered when MkDocs encounters an ontoInk fenced code block. The code block specifies a Turtle data file, an optional SHACL shapes file, and visualization parameters. The plugin then executes the following processing pipeline: 1. RDF Parsing & Classification. The Turtle source is loaded via rdflib Krech and others (2023). Nodes are classified as classes, individuals, literals, or datatypes from rdf:type/rdfs:subClassOf axioms, each receiving a distinct visual notation. 2. SHACL Extraction & Validation. sh:NodeShape instances are traversed; property constraints (sh:path, sh:minCount, sh:maxCount, sh:class) are mapped to graph edges with cardinality badges. The data graph is then validated against the shapes graph using pySHACL Car and Homburg (2020), extracting violations with focus node, property path, severity, and message. 3. OWL Reasoning. Four reasoner backends can be selected per page or request: HermiT Glimm et al. (2014) via owlready2 Lamy (2017) (the default for full OWL DL inference if Java is available); Konclude Steigmiller et al. (2014), an optimized native tableau reasoner, as a Java-free alternative; the same Konclude kernel compiled to WebAssembly via rdf-reasoner-konclude Hanke (2026), which lets OntoInk run OWL DL reasoning entirely in the browser; and, as a lightweight pure-Python fallback, the OWL RL profile Motik et al. (2012). 3.2 Runtime Interaction At page load, the embedded data is decoded, and an interactive graph is initialized. The runtime layer provides these capabilities. Visualization, Navigation and Export. Ontology graphs are rendered using a formal visual notation inspired by established approaches Lohmann et al. (2016); Dudáš et al. (2018). Classes are automatically color-coded by ontology source namespace, enabling immediate identification of cross-ontology dependencies. An Edit Layout panel lets users recolor nodes per declared namespace, change node shapes, and restyle edges, with the customizations carried into the exported image. Node clicks open popups with IRI dereferencing via content negotiation, extracting labels, comments, types, and deprecation status from resolved resources. Inferred triples are shown in a dedicated panel and can be overlaid on the graph as purple dotted edges, so readers can tell asserted statements apart from derived ones. A Validate with Inferences action re-runs SHACL validation over the union of asserted and inferred triples, revealing constraints satisfied only after reasoning. Diagram export is available in publication-ready formats for direct inclusion in publications. SHACL Overlay, Inline Editing and Authoring. SHACL constraints are overlaid as bold dashed edges with cardinality badges [..][min..max], letting engineers inspect structural requirements directly on the graph without reading raw SHACL. An integrated Turtle editor with syntax highlighting enables editing instance data in place; on update, the Turtle is re-parsed, the graph rebuilt, and the data re-validated entirely in the browser. This client-side engine covers the full SHACL Core specification—value-type, cardinality, value-range, string, property-pair, logical, and shape-based constraints. Beyond inspecting existing shapes, OntoInk provides a dedicated SHACL shape editor that lowers the barrier to writing constraints. Users create shapes through a form-based interface instead of editing Turtle directly. They first select a target class and then add property constraints, such as the property path, cardinality, datatype, node kind, or pattern. The corresponding SHACL Turtle representation is generated automatically. 4 Deployment and Use Cases OntoInk has been deployed within the German National Research Data Infrastructure (NFDI) Sure-Vetter et al. (2019), a large-scale initiative for cross-disciplinary research data management. NFDI-MatWerk The MatWerk Knowledge Graph (MatWerk KG)22 2 The MatWerk KG: https://nfdi.fiz-karlsruhe.de/matwerk/, the SHACL shapes are documented using OntoInk and are available at https://nfdi.fiz-karlsruhe.de/matwerk/shapes/ Norouzi et al. (2026) is built on the NFDI MatWerk Ontology (MWO) Beygi Nasrabadi et al. (2026), a BFO-compliant Arp et al. (2015) ontology for research data management in materials science and engineering. On top of MWO, the KG introduces SHACL shapes that enforce data quality. These shapes are documented with OntoInk as interactive diagrams, so that materials scientists unfamiliar with SHACL can inspect the cardinality constraints visually, experiment with instance data in the inline editor, and see the results immediately. Figure 1 shows one of these shapes. The Role Realization shape validates that every role must be realised in at least one process. It requires each instance of BFO_0000023 (role) to carry a BFO_0000054 (realized in) link to the process that brings the role into actuality, so that roles such as principal investigator or peer reviewer are never recorded without the process they participate in. Rendered by OntoInk, the requirement appears as a single constraint edge annotated with its cardinality: a domain expert can edit the instance data in place to watch a violation appear and disappear. The toolbar exposes what OntoInk adds over static documentation: view controls and a fuzzy search that dims non-matching elements rather than removing them. A level-of-detail (LOD) selector governs what is drawn, progressively admitting class hierarchy, individuals and object properties, OWL restrictions and SHACL constraints (from L3), and finally data properties and literals. Analysis panels report graph statistics and run reasoning over one of eight browser- or server-side backends, reporting the inferred triple count, elapsed time, and backend used, overlaying the derived triples on the graph on request, and recording every step of the run log that opens automatically when reasoning fails. Figure 1: The Role Realization SHACL shape of the MatWerk KG, rendered by OntoInk. 5 Conclusion and Demonstration OntoInk brings interactive ontology visualization, SHACL validation, OWL reasoning, and inline Turtle editing together within a documentation-as-code workflow. By treating the Turtle source as the single point of truth, it closes the gap between ontology engineering tools and documentation pipelines, whose general-purpose diagram plugins are not ontology-aware. Reasoning can be performed with HermiT Glimm et al. (2014), native Konclude Steigmiller et al. (2014), or an in-browser WebAssembly port of Konclude Hanke (2026) with an OWL-RL fallback, and instance data is validated against SHACL shapes both at build time with pySHACL and live in the browser. A live demonstration is available at https://ise-fizkarlsruhe.github.io/ontoink/ and includes six interactive example pages covering FOAF, Schema.org, BFO role patterns, IAO textual entities, and OWL reasoning scenarios, together with the browser-based Playground and the visual SHACL shape editor. In future work, we will (i) apply OntoInk to visualize and document developed SHACL shapes and ontology patterns; (i) extend the Edit Layout feature with support for alternative established notations like rendering ontologies in the Graffoo Falco et al. (2014) and Chowlk Chávez-Feria et al. (2022) styles; and (i) improve the visualization of large ontologies through clustering and graph-summarization techniques. Acknowledgements. The authors gratefully acknowledge support from the NFDI-MatWerk consortium as part of the German National Research Data Infrastructure (NFDI) programme. NFDI is financed by the Federal Republic of Germany and the 16 federal states and funded by the Federal Ministry of Education and Research (BMBF) – funding code M532701 / the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) - project number 460247524. aideclaration During the preparation of this work, the author(s) used GPT-4 and Claude Sonnet 4 in order to: Grammar and spelling check. After using these tool(s)/service(s), the author(s) reviewed and edited the content as needed and take(s) full responsibility for the publication’s content. References Alobaid et al. (2019) A. Alobaid, D. Garijo, M. Poveda-Villalón, I. Santana-Perez, A. Fernández-Izquierdo, and O. Corcho Automating ontology engineering support activities with OnToology. Journal of Web Semantics 57, p. 100472. External Links: Document Cited by: Table 1, §2. Arp et al. (2015) R. Arp, B. Smith, and A. D. Spear Building Ontologies with Basic Formal Ontology. MIT Press, Cambridge. Cited by: §4. Barzdins et al. (2010) J. Barzdins, G. Barzdins, K. Cerans, R. Liepins, and A. Sprogis OWLGrEd: a UML style graphical notation and editor for OWL 2. In Proc. of the 7th International Workshop on OWL: Experiences and Directions (OWLED 2010), CEUR-WS, Vol. 614. Cited by: Table 1, §2. Beckett et al. (2014) D. Beckett, T. Berners-Lee, E. Prud’hommeaux, and G. Carothers RDF 1.1 Turtle. Note: W3C Recommendation External Links: Link Cited by: §1. Beygi Nasrabadi et al. (2026) H. Beygi Nasrabadi, E. Norouzi, K. Hubaiev, J. Waitelonis, and H. Sack NFDI MatWerk Ontology (MWO): a BFO-compliant ontology for research data management in materials science and engineering. Advanced Engineering Materials, p. e202502331. External Links: Document Cited by: §4. Car and Homburg (2020) N. J. Car and T. Homburg pySHACL—a Python SHACL validator. External Links: Link Cited by: item 2. Car (2022) N. J. Car pyLODE—an OWL ontology documentation tool. External Links: Link Cited by: Table 1, §2. Chávez-Feria et al. (2022) S. Chávez-Feria, R. García-Castro, and M. Poveda-Villalón Chowlk: from UML-based ontology conceptualizations to OWL. In The Semantic Web (ESWC 2022), LNCS, Vol. 13261, p. 338–352. External Links: Document Cited by: §1, Table 1, §2, §5. Duchateau and Debruyne (2026) J. Duchateau and C. Debruyne OWL2Graffoo: import ontology in diagrams.net with Graffoo. In The Semantic Web: ESWC 2026 Posters and Demos, Note: Accepted at ESWC 2026 Posters and Demos Cited by: Table 1, §2. Dudáš et al. (2018) M. Dudáš, S. Lohmann, V. Svátek, and D. Pavlov Ontology Visualization Methods and Tools: a Survey of the State of the Art. Knowledge Engineering Review 33, p. e10. External Links: Document Cited by: §1, §3.2. Falco et al. (2014) R. Falco, A. Gangemi, S. Peroni, D. Shotton, and F. Vitali Modelling OWL ontologies with Graffoo. In The Semantic Web: ESWC 2014 Satellite Events, LNCS, Vol. 8798, p. 320–325. External Links: Document Cited by: §1, Table 1, §2, §5. Garijo (2017) D. Garijo WIDOCO: a wizard for documenting ontologies. In Proc. of the 16th International Semantic Web Conference (ISWC 2017), LNCS, Vol. 10588, p. 94–102. External Links: Document Cited by: §1, Table 1, §2. Glimm et al. (2014) B. Glimm, I. Horrocks, B. Motik, G. Stoilos, and Z. Wang HermiT: an OWL 2 reasoner. Journal of Automated Reasoning 53 (3), p. 245–269. Cited by: item 3, §5. Gomes et al. (2024) C. A. D. Gomes, G. K. d. S. Q. Monfardini, J. S. Salamon, R. d. S. Sangali, I. S. R. Timoteo, M. Barcellos, and V. E. S. Souza Investigating tool usage in ontology engineering: a survey. In Proc. 17th Seminar on Ontology Research in Brazil (ONTOBRAS 2024), CEUR-WS, Vol. 3905. External Links: Link Cited by: §1. Halilaj et al. (2016) L. Halilaj, N. Petersen, I. Grangel-González, C. Lange, S. Auer, G. Coskun, and S. Lohmann VoCol: an integrated environment to support version-controlled vocabulary development. In Knowledge Engineering and Knowledge Management (EKAW 2016), LNCS, Vol. 10024, p. 303–319. External Links: Document Cited by: §1, Table 1, §2. Hanke (2026) T. Hanke rdf-reasoner-konclude: OWL-DL tableau reasoning in browsers and Node.js via WebAssembly. Note: npm package, https://w.npmjs.com/package/rdf-reasoner-koncludeWASM port of Konclude Steigmiller et al. (2014) Cited by: item 3, §5. Knublauch and Kontokostas (2017) H. Knublauch and D. Kontokostas Shapes constraint language (SHACL). Note: W3C Recommendation External Links: Link Cited by: §1. Krech et al. (2023) D. Krech et al. RDFLib—a Python library for working with RDF. External Links: Link Cited by: item 1. Lamy (2017) J. Lamy Owlready: ontology-oriented programming in Python with automatic classification and high level constructs for biomedical ontologies. Artificial Intelligence in Medicine 80, p. 11–28. Cited by: item 3. Lieber et al. (2022) S. Lieber, B. De Meester, P. Heyvaert, F. Bruckmann, R. Wambacq, E. Mannens, R. Verborgh, and A. Dimou Visual notations for viewing RDF constraints with UnSHACLed. Semantic Web 13 (5), p. 757–792. External Links: Document Cited by: Table 1, §2. Lohmann et al. (2015) S. Lohmann, V. Link, E. Marber, and S. Negru WebVOWL: web-based visualization of ontologies. In Knowledge Engineering and Knowledge Management, LNCS, Vol. 8982, p. 154–158. External Links: Document Cited by: §1, Table 1, §2. Lohmann et al. (2016) S. Lohmann, S. Negru, F. Haag, and T. Ertl Visualizing ontologies with VOWL. Semantic Web 7 (4), p. 399–419. External Links: Document Cited by: §2, §3.2. Mermaid Contributors (2014) Mermaid: generation of diagrams and flowcharts from text in a similar manner as markdown External Links: Link Cited by: §1. Motik et al. (2012) B. Motik, B. C. Grau, I. Horrocks, Z. Wu, A. Fokoue, and C. Lutz OWL 2 web ontology language profiles (second edition). Note: W3C Recommendation External Links: Link Cited by: item 3. Mouromtsev et al. (2025) D. Mouromtsev et al. Reactodia: a React-based library for interactive visualization and editing of RDF/OWL knowledge graphs. Note: Open-source library, https://reactodia.github.ioSuccessor of OntoDia Mouromtsev et al. (2015) Cited by: Table 1, §2. Mouromtsev et al. (2015) D. Mouromtsev, D. Pavlov, Y. Emelyanov, A. Morozov, D. Razdyakonov, and M. Galkin The simple web-based tool for visualization and sharing of semantic data and ontologies. In Proc. ISWC 2015 Posters & Demos, CEUR-WS, Vol. 1486. Cited by: Table 1, §2, 25. Musen (2015) M. A. Musen The Protégé project: a look back and a look forward. AI Matters 1 (4), p. 4–12. External Links: Document Cited by: Table 1. Norouzi et al. (2026) E. Norouzi, H. Beygi Nasrabadi, G. Singh, A. Algergawy, J. Waitelonis, and H. Sack MSE-KG v2.1.1 – materials science and engineering knowledge graph. Zenodo. External Links: Document Cited by: §4. Peroni et al. (2012) S. Peroni, D. Shotton, and F. Vitali The live OWL documentation environment: a tool for the automatic generation of ontology documentation. In Proc. EKAW 2012, LNCS, Vol. 7603, p. 398–412. External Links: Document Cited by: §2. Steigmiller et al. (2014) A. Steigmiller, T. Liebig, and B. Glimm Konclude: system description. Journal of Web Semantics 27–28, p. 78–85. External Links: Document Cited by: item 3, §5, 16. Sure-Vetter et al. (2019) Y. Sure-Vetter, C. Goble, and M. Nyström Enabling the National Research Data Infrastructure (NFDI). In GI-Jahrestagung, p. 2275–2276. Cited by: §4. Terrastruct (2022) D2: a modern diagram scripting language Note: GitHub repository. Accessed: 2026-07-20 External Links: Link Cited by: §1. Wiens et al. (2018) V. Wiens, S. Lohmann, and S. Auer WebVOWL editor: device-independent visual ontology modeling. In Proc. ISWC 2018 Posters & Demonstrations, CEUR-WS, Vol. 2180. Cited by: Table 1.

Instant research discovery