Visualizing OWL and RDF: Advancing Ontology Representation for

Enhanced Semantic Clarity and Communication

Giulia Biagioni

Unit ISP, Netherlands Organisation for Applied Scientiﬁc Research (TNO), New Babylon, Anna van Buerenplein 41a,

Den Haag, The Netherlands

Keywords:

Ontology Engineering, Visualization Tools, Standardized Visual Representation, RDF Semantics,

OWL Semantics.

Abstract:

This position paper advocates for the development of advanced visualization tools speciﬁcally designed to rep-

resent the full range of expressiveness conveyed by the vocabulary terms of OWL and RDFs. It highlights the

urgent need for a standardized way to visually refer to these vocabulary terms, addressing current challenges

in ontology visualization. The paper outlines the signiﬁcant beneﬁts that such tools and initiatives will deliver,

including enhanced clarity, improved communication among stakeholders, and more efﬁcient management

of semantic data. By standardizing visual representations, the paper argues for a more intuitive and accessi-

ble approach to interacting with complex semantic structures, ultimately facilitating better understanding and

broader adoption of semantic web technologies.

1 INTRODUCTION

In the landscape of European Union initiatives, in-

cluding the establishment of data spaces and Inter-

net of Things (IoT) driven frameworks for smart en-

vironments, ontologies have emerged as essential in-

struments (Corcho and Simper, 2022)(Commission,

2022). These frameworks are instrumental in stan-

dardizing data across a myriad of sectors, enhanc-

ing interoperability, ensuring regulatory compliance,

and augmenting the functionality and integration of

smart technologies. This integration fosters interop-

erable and efﬁcient communication among devices.

Despite the widespread acknowledgment of their util-

ity and the broad adoption across various domains, a

signiﬁcant challenge persists: the lack of efﬁcient vi-

sualization tools for examining the logical structures

of ontologies deﬁned through RDFs (Resource De-

scription Framework) and OWL (Web Ontology Lan-

guage) standardized semantics

Current visualization tools fall short in enabling

users to effectively and thoroughly explore the intri-

cacies of ontology structures, encompassing their ax-

ioms, rules, and taxonomical classiﬁcations (Dud

https://orcid.org/0000-0002-9005-7945

For RDFs and OWL semantics cf. respectively: https:

//www.w3.org/TR/rdf-schema/ and https://www.w3.org/

TR/owl-guide/

et al., 2018). This gap in technological capability

signiﬁcantly impedes users’ ability to conduct com-

prehensive inspections, which are critical to verify-

ing the consistency and accuracy of semantic con-

structions within ontologies. Moreover, this situa-

tion is compounded by the limitations inherent in ex-

isting RDFs and OWL-based reasoners, which often

struggle to identify contradictions or process infer-

ences with the necessary precision for ensuring se-

mantic consistency. Adding to these challenges is the

lack of a standardized approach for visually represent-

ing the vocabulary and symbols associated with OWL

and RDFS. This absence of a uniform visual language

hinders effective communication and comprehension

among stakeholders who interact with these semantic

systems.

There is a pressing need not only for improved

visualization tools that can adeptly handle the com-

plexity of ontological structures but also for the de-

velopment of a cohesive visual language that can

standardize the representation of semantic constructs.

Such advancements would greatly enhance clarity and

usability, enabling users to more effectively engage

with and utilize semantic technologies in their respec-

tive ﬁelds. This position paper will delve into these

challenges and propose strategic pathways to address

these critical gaps. The content of this paper is pre-

sented by respecting the following order. Section 1

Biagioni, G.

Visualizing OWL and RDF: Advancing Ontology Representation for Enhanced Semantic Clarity and Communication.

DOI: 10.5220/0012855900003756

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 13th International Conference on Data Science, Technology and Applications (DATA 2024), pages 571-576

ISBN: 978-989-758-707-8; ISSN: 2184-285X

571

delineates the introduction. Section 2 addresses the

current state of ontology visualization tools and the

challenges faced by ontology engineers. Section 3 in-

vestigates the shortcomings of reasoners in analyzing

ontological structures and provide support to ontol-

ogy engineers in assessing the logical consistency of

the ontology. Section 4 discusses how semiotic prin-

ciples can enhance ontology visualization by creat-

ing intuitive and meaningful representations that fa-

cilitate better communication and understanding. The

conclusion emphasizes the critical need for integrat-

ing semiotic principles into visualization tools as the

semantic web grows, advocating for both the develop-

ment of visualization tools targeting the semantics of

OWL and RDFs and the creation of a standardized vi-

sual language (i.e set of symbols used in visualization

tools for ontologies) to refer to them.

2 STATUS & CHALLENGES

In the ﬁeld of the semantic web, no visualization tool

or method has yet been universally adopted as the

standard, according to Dudas and colleagues (Dud

et al., 2018). This lack of a standard can be attributed

primarily to two main reasons. Firstly, there is no

’one-size-ﬁts-all’ solution that caters to the varied re-

quirements of different users. Secondly, there has

been minimal progress in this area since 2007.

Dudas et al. (Dud

as et al., 2018) identify four

primary high-level use cases for ontology visualiza-

tion tools: editing, learning, inspecting, and sharing.

Visual editing involves generating an ontology script

automatically from a diagram, a feature particularly

useful for users unfamiliar with RDF(S) and OWL

syntax. Visualization tools also play a crucial role

in learning and inspecting ontology structures. These

activities are primarily aimed at ontology engineers

who need to identify and rectify semantic errors, in-

consistencies, and contradictions. Finally, sharing

knowledge representations visually provides an ef-

fective means to quickly communicate the scope of

ontologies, thereby enhancing understanding and col-

laboration.

Visualization tools offer a range of beneﬁts that

can signiﬁcantly enhance information processing and

management (Bernasconi et al., 2023). By increas-

ing cognitive memory and optimizing resource pro-

cessing, these tools make it easier and faster to ac-

cess and assimilate information (Skulmowski and Xu,

2022). They simplify the search for information and

enhance the ability to detect patterns, thus facilitating

quicker and more effective decision-making. Visual-

ization tools also support perceptual inference oper-

ations and improve attention mechanisms, which are

crucial for monitoring complex systems. Addition-

ally, these tools encode information in a format that is

easy to manipulate, allowing users to handle and ana-

lyze large datasets efﬁciently. They provide a com-

prehensive overview while enabling detailed explo-

ration, helping users to maintain a broad perspective

even as they delve into speciﬁc data points. Further-

more, visualization tools are invaluable for tracking

various elements and activities, and they excel at pro-

ducing abstract representations of scenarios by selec-

tively omitting and emphasizing certain pieces of in-

formation, thereby simplifying complex situations.

The visualization of ontologies, encompasses a di-

verse array of methods each suited to speciﬁc use

cases, including indented lists, node-link visualiza-

tions, force-directed layouts, tree layouts, radial lay-

outs, circle layouts, inverted radial tree layouts, Euler

diagrams, treemaps, 2.5D visualizations, and 3D vi-

sualizations. Each method caters to distinct aspects

of ontology visualization, from simple indented lists

that show hierarchical relationships in a straightfor-

ward, text-based format to more complex 3D visual-

izations that provide a spatial representation of data

but may struggle with the translation to 2D screens.

Additional features enhancing these visualizations in-

clude radar views for a minimap perspective, graphi-

cal and semantic zooming to adjust detail, entity fo-

cus to highlight speciﬁc data points, and history track-

ing for navigation. Advanced techniques like concept

clustering and semantic zooming allow for dynamic

interaction with the ontology, emphasizing the visual-

ization’s adaptability to user needs and the complex-

ity of the information displayed. Ultimately, visual-

ization tools for ontologies serve two primary pur-

poses: they can display data records structured ac-

cording to the ontologies, or they can visualize the

semantic constructs in OWL and RDFs that form the

foundation of the ontologies themselves. The latter

category of tools is crucial for illustrating how on-

tologies are built, revealing the composition and in-

teractions of various semantic elements that enable

machines to comprehend and process the meaning of

information and data, and infer new data.

Regarding specialized visualization tools de-

signed for displaying OWL and RDFS constructs us-

ing ’signs’ or ’symbols’, two notable examples are

Chowlk (Ch

avez-Feria et al., 2022) and WebVOWL

(Lohmann et al., 2017). Chowlk

enables users to

edit ontologies through a visualization interface, mak-

ing it an invaluable resource for those needing to

develop or modify ontologies while assessing their

Chowlk is accessible at: https://chowlk.linkeddata.es/

notation.html

DATA 2024 - 13th International Conference on Data Science, Technology and Applications

572

structure and semantics. Conversely, WebVOWL

focuses exclusively on inspection, offering a platform

for users to examine and understand the relationships

and structures within an ontology, though it lacks edit-

ing capabilities. Despite their signiﬁcant contribu-

tions to enhancing the accessibility and usability of

ontological data, these tools also face speciﬁc limita-

tions. Chowlk, for example, does not permit the up-

loading and visualization of existing ontologies, lim-

iting its applicability. Meanwhile, WebVOWL, while

providing a visual insight into ontology semantics,

falls short in effectively revealing the detailed logi-

cal structure of ontologies, which can hinder a thor-

ough analysis of their construction. Moreover, both

tools do not support the visualization of all RDFs and

OWL constructs.

Figure 1: Example visualization for the semantic

OWL construct (owl:minQualiﬁedCardinality) adopted by

Chowlk. The image has been sourced from https : / /

chowlk.linkeddata.es/notation.html.

Figure 2: Example visualization for semantics OWL and

RDFs constructs adopted by WebVOWL. The image has

been sourced from https://service.tib.eu/webvowl/.

The lack of visualization tools capable of allow-

ing users to fully visually inspect the axiomatic and

taxonomic structures of vocabularies poses a signiﬁ-

cant challenge. This problem is compounded by the

absence of a standardized method for visually repre-

senting OWL and RDFs constructs. Such limitations

create a substantial barrier for those working to de-

velop, reﬁne, or simply understand the intricate rela-

tionships and rules within these semantic frameworks.

The ability to visually navigate these structures is cru-

cial for effective analysis and veriﬁcation of ontologi-

cal consistency, which underpins many advanced data

processing and AI systems. Moreover, the gaps in vi-

sualization capabilities are particularly problematic in

accessible at: https : / / github.com / VisualDataWeb /

WebVOWL

light of the limitations of open-source reasoners, as it

will be further explained in the subsequent section.

3 REASONERS’ LIMITS

Reasoners are powerful tools used in the semantic

web to process and infer logical consequences from

a set of ontologies. Inference is the process by which

these tools deduce new knowledge from the facts al-

ready deﬁned in the ontology. For example, if an on-

tology states that all mammals are warm-blooded and

that whales are mammals, a reasoner can infer that

whales are warm-blooded. This ability to derive new

facts extends the utility of ontologies, enabling more

complex queries and enhanced data interaction.

Contradictions occur when conﬂicting informa-

tion is deduced or entered into the ontology, such as

stating both that no birds are mammals and that some

birds are mammals. Effective reasoners are designed

to detect these contradictions, which are critical for

maintaining the integrity and reliability of data. How-

ever, detecting contradictions does not always happen

automatically. In complex ontologies where multiple

relationships and properties interact in subtle ways,

contradictions might be less apparent and can be over-

looked by reasoners, especially if the reasoner is not

fully compliant with all rules and nuances of the on-

tology language or is limited by performance con-

straints.

According to Gomez-Perez et al. (Gomez-Perez,

1994), (Gomez-Perez et al., 2001), (Gomez-Perez

et al., 1999) and Fahad et al. (Fahad et al., 2007), (Fa-

had and Qadir, na), inconsistencies and contradictions

in ontologies typically arise from three main types of

errors: Circluatory Errors, Partition Errors and Se-

mantic Inconsistency Errors. Circulatory Errors oc-

cur when a class or property is wrongly deﬁned as

a subclass or superclass (or subproperty, superprop-

erty) of itself, creating a loop in the ontology’s hier-

archy. Partition Errors happen when an entity, such

as a class, property, or instance, is incorrectly clas-

siﬁed under multiple disjoint categories, often dur-

ing the process of breaking down a concept into sub-

concepts or when dealing with exhaustive decompo-

sition where an external instance does not ﬁt any de-

ﬁned category. Lastly, Semantic Inconsistency Errors

arise from incorrect hierarchical classiﬁcations, such

as placing a concept as a subclass under an incompat-

ible superclass. This type of error usually stems from

three situations: subclasses representing a broader do-

main than their superclass (i.e. weaker domain spec-

iﬁed by subclass error), subclasses introducing char-

acteristics that contradict their superclass’s traits (i.e.

Visualizing OWL and RDF: Advancing Ontology Representation for Enhanced Semantic Clarity and Communication

573

domain breach speciﬁed by subclass error), and sub-

classes being placed under a concept from a disjoint

domain (i.e disjoint domain speciﬁed by subclass er-

ror).

In 2008, Fahad et al (Fahad et al., 2008). con-

ducted a test which revealed that a number of reason-

ers were unable to detect partition errors and semantic

inconsistency errors. Recently, I replicated this exact

experiment using a selection of well-regarded reason-

ers to assess their current capabilities. Among these

were HermiT 1.4.3.456, known for its strict adher-

ence to OWL 2 DL ontologies, and Pellet 2.2, still

relevant for its extensive OWL 2 DL support. I also

tested Pellet Incremental to evaluate its efﬁcacy in

dynamic environments where ontologies frequently

change, allowing for real-time updates without the

need for full reprocessing. Additionally, Elk, valued

for its speed in handling EL++ ontologies despite po-

tential stability issues, and FaCT++, recognized for its

performance albeit with installation challenges, were

included in the test. The ontology used was the same

as described by Fahad et al.(Fahad et al., 2008), and

the reasoners were operated within the Protege frame-

work. The following results were attained.

In the experiment replicating the tests originally

performed by Fahad et al.(Fahad et al., 2008), I found

that the current selection of reasoners still struggled

with detecting speciﬁc types of semantic inconsis-

tency errors. Notably, they failed to identify issues

such as weaker domain speciﬁed by subclass error.

For example, an error was not detected in a scenario

where ’Owner owns some PassengerVehicle’ was cre-

ated as a subclass of the concept ’Owner4Vehicle

owns exactly 4 PassengerVehicle’. Additionally, the

Domain Breach Speciﬁed by Subclass Error went un-

detected, such as in the case where ’OwnerManyVe-

hicle’, deﬁned as owning more than 3 but less than 8

PassengerVehicles, was incorrectly subclassed under

’Owner of exactly 4 Vehicles’.

Moreover, the reasoners did not detect violations

of the OWL propertyDisjointWith vocabulary term,

which speciﬁes that two properties should never relate

the same pair of resources. Instances where properties

supposedly disjoint still related the same entities were

overlooked by reasoners.

The experiment also revealed shortcomings in the

reasoners’ ability to infer new information based on

conceptual distances or hierarchies extending beyond

an arc distance of +10 from the initial concept. Fur-

thermore, Circulatory Errors, which lead to circular

reasoning and thus inconsistencies, were not recog-

nized by the reasoners. These results highlight per-

sistent gaps in the capabilities of ontology reasoners,

underscoring the need for ongoing improvements in

their algorithms to handle complex logical constructs

more effectively. Ultimately, these ﬁndings empha-

size that reasoners cannot solely be relied upon by

ontology engineers during the conceptualization pro-

cess or when assessing the consistency and efﬁciency

of their models. This underscores the necessity for

additional tools to aid in the development and veriﬁ-

cation of ontologies. Visualization tools, in particular,

can offer substantial support in this regard, provided

there is a standardized and uniform way used to visu-

ally represent and refer to OWL and RDFS constructs.

By integrating such visualization tools, ontology en-

gineers can gain better insights into the structure and

interrelations within their ontologies, making it eas-

ier to spot inconsistencies and inefﬁciencies that rea-

soners might miss. This dual approach of using both

reasoners and visualization tools can signiﬁcantly en-

hance the robustness and utility of semantic data mod-

eling.

Building on the identiﬁed limitations of ontology

reasoners and the potential of visualization tools to

enhance the conceptualization and validation process

in semantic data modeling, it becomes apparent that

integrating semiotics (i.e. the study of signs and sym-

bols and their use or interpretation) into ontology en-

gineering could provide further advantages. Just as

reasoners parse and interpret data based on logical

constructs, visualization tools can use semiotic prin-

ciples to encode and represent complex information,

making abstract concepts more accessible and inter-

pretable as it will be discussed in the next part.

4 THE SEMIOTICS OF

VISUALIZATION TOOLS

Semiotics, the study of signs and symbols and their

interpretation, serves as the foundational framework

of semantics, crucial for building the semantic web.

Signs encompass a variety of forms including words,

images, objects, and even smells, which acquire

meaning only when humans assign signiﬁcance to

them. This ﬁeld is heavily inﬂuenced by the theories

of Ferdinand de Saussure and Charles Sanders Peirce,

as highlighted by Vickers et al. (Vickers et al., 2013).

Saussure’s model explains a sign as consisting of a

signiﬁer (the form the sign takes, such as a word or

sound) and the signiﬁed (the concept it represents).

For example, the sound pattern /cat/ signiﬁes the ani-

mal cat. Conversely, Peirce’s model introduces a tri-

adic relationship involving the object (the entity being

represented), the representamen (the physical form of

the sign), and the interpretant (the understanding at-

tributed to the sign by observers).

DATA 2024 - 13th International Conference on Data Science, Technology and Applications

574

Applying these principles to ontology visualiza-

tion allows for the use of diverse signs to effectively

demonstrate how different ontologies are structured,

thereby enabling various types of structural analyses.

Each sign or symbol in a visualization could be delib-

erately designed to represent speciﬁc OWL and RDFS

vocabulary terms. This method not only standard-

izes visual representations but also assigns speciﬁc,

agreed-upon meanings to these signs. For example,

different shapes, colors, or icons might be used to de-

note classes, properties, subclasses, and other onto-

logical elements, transforming a visual diagram into

a readable, meaningful map of an ontology.

This semiotic encoding bridges the gap between

the complex data managed by reasoners and the hu-

man understanding necessary for efﬁcient ontology

engineering. It allows engineers to quickly grasp the

ontology’s structure and logic, facilitating the identi-

ﬁcation of relationships, hierarchies, and potential in-

consistencies without needing to parse complex codes

or logical expressions. Ultimately, integrating the

computational power of reasoners with the intuitive

clarity of semiotically-enhanced visualization tools

enables ontology engineers to achieve a deeper, more

nuanced understanding of their data models.

The creation of a tool designed to visualize stan-

dardized symbols representing vocabulary terms of

OWL and RDFS would bring signiﬁcant beneﬁts to

the ﬁeld of ontology engineering. Such a tool would

greatly enhance the clarity and accessibility of com-

plex ontological structures, making it easier for en-

gineers and stakeholders to grasp and interact with

the semantic layers underlying their data. Standard-

ized visual representations would facilitate quicker

and more accurate interpretation of relationships and

hierarchies within ontologies, reducing the cognitive

load on users and speeding up both the design and

troubleshooting processes. Additionally, by provid-

ing a universal language of visual symbols, this tool

would promote better communication and collabora-

tion across diverse teams and disciplines, ensuring

that ontologies are not only consistently understood

but also more effectively integrated into broader sys-

tems. This could lead to advancements in data inter-

operability and more seamless integration of seman-

tic web technologies across different platforms and

applications, ultimately driving innovation and efﬁ-

ciency in the use and management of knowledge.

5 CONCLUSION

The integration of semiotic principles into ontology

visualization tools is not merely an enhancement but a

necessity. As the Semantic Web continues to grow in

both size and signiﬁcance, the ability to effectively vi-

sualize ontological structures in a semantically mean-

ingful way becomes critical. Tools that can bridge the

gap between complex semantic structures and user-

friendly visual representations will play a pivotal role

in the advancement of knowledge engineering. This

endeavor not only supports the technical development

of ontologies but also facilitates a broader understand-

ing and adoption of semantic technologies across var-

ious sectors.

In this evolving landscape, the development and

standardization of visualization tools capable of ac-

curately representing OWL and RDFs constructs are

paramount. By creating systems that visually encode

ontology semantics, we enable more intuitive interac-

tions with data, enhance the capacity for insight, and

signiﬁcantly improve the accuracy of ontology engi-

neering tasks. This approach not only aids ontology

engineers in managing the inherent complexities of

semantic frameworks but also empowers stakeholders

across different domains to engage with and leverage

the full potential of semantic technologies.

Moreover, these visualization tools serve as a crit-

ical complement to reasoners, addressing the cur-

rent limitations in detecting inconsistencies and in-

ferring new knowledge. By integrating robust visu-

alization technologies with advanced reasoning capa-

bilities, we can forge a more comprehensive toolkit

for ontology engineers, enhancing both the develop-

ment and application of ontologies within the digital

ecosystems.

Thus, as the advocacy for the advancement of se-

mantic web technologies continues, the focus must

also shift towards reﬁning the tools that facilitate this

advancement. Establishing a standard for visual rep-

resentations of semantic data, paired with ongoing en-

hancements to reasoning technologies and visualiza-

tion tools, will not only address the current challenges

but also pave the way for future innovations in this

critical area of technology.

ACKNOWLEDGEMENTS

This research activity was conducted as part of the

Norm Engineering Program at TNO, funded by the

Dutch Ministry of the Interior and Kingdom Rela-

tions. I would also like to extend my gratitude to my

colleague Ioannis Tolios for his valuable feedback on

the paper.

Visualizing OWL and RDF: Advancing Ontology Representation for Enhanced Semantic Clarity and Communication

575

REFERENCES

Bernasconi, E., Ceriani, M., Di Pierro, D., Ferilli, S., and

Redavid, D. (2023). Linked data interfaces: A survey.

Information, 14(9):483.

avez-Feria, S., Garc

ıa-Castro, R., and Poveda-Villal

on,

M. (2022). Chowlk: From uml-based ontology con-

ceptualizations to owl. In Proceedings of the 19th Ex-

tended Semantic Web Conference (ESWC 2022), page

Conference speciﬁc pagination. Accessed: 2024-04-

18.

Commission, E. (2022). Moving towards a euro-

pean data space: New eu law for data-sharing.

https://data.europa.eu/en/news-events/news/moving-

towards - european - data - space - new - eu - law - data -

sharing. Last accessed 2024/04/18.

Corcho, O. and Simper, E. (2022). data.europa.eu and the

european common data spaces, a report on challenges

and opportunitie.

Dud

as, M., Lohmann, S., Sv

atek, V., and Pavlov, D. (2018).

Ontology visualization methods and tools: A survey

of the state of the art. The Knowledge Engineering

Review, 33.

Fahad, M., Qadir, M., and Noshairwan, W. (2007). Seman-

tic inconsistency errors in ontologies. In Proceedings

of the IEEE GRC 2007, pages 283–286, Silicon Val-

ley, USA. IEEE Computer Society.

Fahad, M. and Qadir, M. A. (n.a). A framework for ontol-

ogy evaluation. In Proceedings of the Workshop at

CEUR Workshop Proceedings, Mohammad Ali Jin-

nah University, Islamabad, Pakistan. CEUR Work-

shop Proceedings.

Fahad, M., Qadir, M. A., and Shah, S. (2008). Evaluation of

ontologies and dl reasoners. In Intelligent Information

Processing IV, volume 288, pages 17–27. Springer.

Gomez-Perez, A. (1994). Some ideas and examples to eval-

uate ontologies. Technical report, Knowledge Sys-

tems Laboratory, Stanford University, Stanford, CA.

Gomez-Perez, A. et al. (1999). Evaluation of taxonomic

knowledge on ontologies and knowledge-based sys-

tems. In Proceedings of the International Workshop

on Knowledge Acquisition, Modeling and Manage-

ment.

Gomez-Perez, A., Lopez, M., and Garcia, O. (2001). On-

tological Engineering: With Examples from the Areas

of Knowledge Management, E-Commerce and the Se-

mantic Web. Springer.

Lohmann, S., Haag, F., Ertl, T., and Negru, S. (2017). Visu-

alizing ontologies with vowl. Semantic Web Journal.

Accessed: 2024-04-18.

Skulmowski, A. and Xu, K. M. (2022). Understanding cog-

nitive load in digital and online learning: A new per-

spective on extraneous cognitive load. Educational

Psychology Review, 34:171–196.

Vickers, P. et al. (2013). Understanding visualization: A

formal approach using category theory and semiotics.

IEEE Transactions on Visualization and Computer

Graphics.

DATA 2024 - 13th International Conference on Data Science, Technology and Applications

576