REQUIREMENTS

FOR INTERACTIVE ONTOLOGY VISUALIZATION

Using Hypertree +2.5D Visualization for Exploring

Relationships between Concepts

Isabel Cristina Siqueira da Silva

1,2

and Carla Maria Dal Sasso Freitas

Instituto de Informática, UFRGS, Av. Bento Gonçalves, 9500, Porto Alegre, RS, Brazil

Faculdade de Informática, Centro Universitário Ritter dos Reis, UNIRITTER

Rua Orfanotrófio, 555, Porto Alegre, RS, Brazil

Keywords: Visualization, Ontology, Interaction.

Abstract: Ontologies are used for sharing among people or software agents the common understanding of the

information structure in a certain domain. Usually, ontologies are represented as static 2D graphs where the

relationships are displayed as edges, which often overlap and cause cognitive overload. Three-dimensional

representations can also lead to confusion due to occlusion. Moreover, as the ontology grows, incorporating

new concepts (and their relationships) increases the visualization complexity either in 2D or in 3D. In this

paper, we present a study about the requirements of visualization and interaction with ontologies. In order to

do that, we interviewed with four experts on ontology creation and use. From the results, we propose the

design of a 2.5D visualization tool for exploring relationships between ontology concepts.

1 INTRODUCTION

There is a gradual increase of information available

and efficient methods for information retrieval are

necessary in order to allow interoperability and

cooperation between several databases. Data

semantics is the more traditional approach for data

integration because it focuses on the relationship

between data. As such, ontologies define concepts

and ensure interoperability between systems. In his

work, Sowa (2005) points out that ontology is the

study of the categories of things that exist or may

exist in some domain, i.e., it is a catalogue of the

types of things that are assumed to exist in a domain

of interest D from the perspective of a person who

uses a language L for the purpose of talking about D.

According to Gruber (1996), ontology is a formal

and explicit specification of a conceptualization.

Noy and McGuiness (2001) discuss that ontologies

allow sharing the common understanding of the

structure of information among people or software

agents. Ontologies separate domain knowledge from

the operational knowledge, make domain

assumptions explicit and enable reuse.

However, due to the specificities of the concepts

expressed in ontologies, the analysis of individual

relationships is complex. Thus, interactive ontology

visualizations need to be efficient and allow rapid

comprehension of concepts and relationships.

Katifori (2007) confirms that it is not simple to

create a visualization that displays effectively all the

information, and, at the same time, allows the user to

perform easily various operations on the ontology.

Then, the challenge is to define the best way to

represent relationships between categorized concepts

mainly because each concept can have a number of

related attributes.

This work presents requirements analysis for

visualization and interaction in tools aiming at

creating, manipulating and exploring ontologies. We

conducted interviews with users who work with

ontologies and conceptual modelling. From these

results, we present an initial design of a 2.5D

ontology visualization method that aims at

systematizing and transmitting knowledge more

efficiently. The text is organized as follows. Section

2 discusses related work. Section 3 presents the user

interviews and points out requirements for ontology

visualization tools. Section 4 presents our proposal

for ontology visualization. Results are discussed and

final comments are drawn in Section 5.

242

Siqueira da Silva I. and Dal Sasso Freitas C..

REQUIREMENTS FOR INTERACTIVE ONTOLOGY VISUALIZATION - Using Hypertree +2.5D Visualization for Exploring Relationships between

Concepts .

DOI: 10.5220/0003372402420248

In Proceedings of the International Conference on Imaging Theory and Applications and International Conference on Information Visualization Theory

and Applications (IVAPP-2011), pages 242-248

ISBN: 978-989-8425-46-1

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

2 RELATED WORKS

Different authors propose alternatives for

visualization and interaction with ontologies.

Katifori (2007) discusses different techniques

that could be adapted for ontology representation,

such as indented lists, trees and graphs, zooming,

space subdivision (treemaps, information slices),

focus+context and landscapes. Besides that, tools for

ontology visualization and interaction are discussed.

Fluit et al. (2005) present the cluster map technique

as a simple and intuitive method for complex

ontologies visualization.

The OntoSphere tool (Bosca et al., 2005) uses

two techniques - 3D and focus+context – for

providing overview and details according to user

needs. Baehrecke (2004) and Babaria (2004) are

proposing the use of treemaps to visualize GO (Gene

Ontologies Consortium). In a treemap, colour, size

and grouping are used in order to facilitate user

interaction and information extraction.

Protégé (Noy et al., 2000) is the common

software used for the creation and visualization of

ontologies. Protégé’s main visualization for the

ontology hierarchy is a tree view (Class Browser).

However, different visualization techniques have

been proposed: Katifori (2008) presents a

comparative study of four visualization techniques

available in past versions of Protégé: Class Browser,

Jambalaya (discontinued), TGVizTab (discontinued)

and OntoViz. The information retrieval provided by

these tools was also evaluated.

Lanzenberger (2009) discusses the importance of

ontology visualization based on graphs, as well as

tools for mapping and alignment of ontologies, and

views, which employ different structures of graphs

for ontologies visualization. These techniques are

compared in order to point out their advantages and

disadvantages.

Catenazzi et al. (2009) presents a study about

tools for ontologies visualization and proposes the

OWLeasyViz tool. It combines textual and graphical

representations for displaying the class hierarchy,

relationships and data properties. Interaction

techniques such as zooming, filtering and search are

available. Kriglstein and Wallner (2010) present

Knoocks, a visualization tool focused on the

interconnections within the ontology and the

instances. This tool employs the overview + details

approach.

The works by Samper et al. (2008) and Amaral

(2008) address semantics aspects. Amaral (2008)

proposes a semantics-based framework for

visualizing descriptions of concepts in OWL. The

framework aims at allowing users to obtain deep

insights about the meaning of such descriptions,

thereby preventing design errors or misconceptions.

Icons and symbols are used in diagrams to

characterize classes that represent concepts

descriptions. One can combine information

visualization techniques, as in the work by Schevers

et al. (2008), where the user interacts with the

ontology in the Protégé tool. Classes representing

spatial information (like polygons, points, etc.) are

presented in a second graphical interface that is used

to mimic the functionality of a GIS (Geographic

Information System).

The paper by Kriglstein (2009) presents a survey

about users’ attitudes and expectations regarding

ontology visualization, pointing out some

requirements.

3 REQUIREMENTS

FOR INTERACTIVE

VISUALIZATION

OF ONTOLOGIES

The main reason for the use of ontologies is to

provide efficient information retrieval, identifying

non-explicit relationships between data. Thus,

information categorization in ontology modelling is

very important.

Ontologies tend to grow, incorporating new

concepts and relationships, therefore increasing the

visualization complexity. Static graphs, commonly

used for ontology representation (Figure 1), are not

the best alternative for such visualization. Figure 1,

for example, presents two visualizations of the

concepts of an ontology hierarchy built with

Protégé: tree view (Class Browser) and static graph

(OWLViz). The static graph may interfere in the

user's perception about the universe represented in

the ontology because the display of relationships

generates overlapping edges. Likewise, tree view

shows the hierarchy of concepts but not their

relationships. The solution for these problems may

be the use of different information visualization

techniques.

Information visualization and concepts of human

computer interaction can optimize the

comprehension of ontologies. The searched

information can be placed in focus, distinguishing it

from the unnecessary information and facilitating

the understanding of correlated data.

Designers of visualization systems should

consider two main issues: the mapping of

REQUIREMENTS FOR INTERACTIVE ONTOLOGY VISUALIZATION

- Using Hypertree +2.5D Visualization for Exploring Relationships between Concepts

243

Figure 1: Visualization in Protégé: tree view (left) and static graph (right).

information for a graphical representation in order to

facilitate its interpretation by the users, and means to

limit the amount of information that users receive,

while keeping them "aware" of the total information

space and reducing cognitive effort.

As presented in section 2, many studies have

addressed the issue of the importance of ontology

visualization in creation, manipulation and inference

processes. Different visualization methods have

been proposed, but there are still many gaps to be

filled by efficient methods of visualization and

interaction. The study presented in this paper takes

into account the results already discussed by other

authors, like Katifiori (2007) and Kriglstein (2009),

adding new ideas obtained from the users

interviews.

3.1 Interviews with Users

This study started with interviews with four users,

from the Graduate Program in Computer Science at

UFRGS. They work with ontologies daily and can

be considered experts in the creation and

manipulation of ontologies. From these interviews,

we confirmed some requirements pointed out by

Kriglstein (2009), and list other necessary

requirements for ontology visualization tools.

Three experts are within the Intelligent Database

(IDB) group, whose research focus is knowledge

engineering, case-based reasoning, document

retrieval, information management, and intelligent

databases. Another expert belongs to a group

studying quality of information on the web and

recommendation systems.

Due to the low number of participants,

quantitative measurements were not taken, but the

qualitative notes are very interesting as indicated by

Nielsen (1994). The following questions were posed

to the experts:

1) When an ontology is created, which aspects

could be improved with visualization?

2) After the ontology was created, which

information is searched more often and how this

information could be displayed in order to make

understanding more efficient?

3) When and why a visualization is better than

another?

For question (1), users responded that the main focus

is on the elements that define the structure of the

ontology. These elements refer to the relations

between class and subclasses, between classes, and

between the instances of classes. An ideal

visualization tool should focus on the ontology

kernel (question 3).

During the process of ontology development,

users want to visualize different aspects of specific

demands that arise in a certain stage of development.

Thus, display features could have privileged access

at certain points, for example, checking the range of

an object property. However, the development of

ontologies is still a traditional activity, with no

defined workflow and directly influenced by the

domain.

Another important aspect related to question (1)

is the visualization of the ontology validation

generated by inference processes. Displaying errors

can (should) be improved, with the indication of

correction.

IVAPP 2011 - International Conference on Information Visualization Theory and Applications

244

In relation to question (2), we must consider that

ontologies can encapsulate a large amount of

information (hundreds of thousands of classes and

relationships, for example). Moreover, this large

volume of information can be segmented into

several distinct types (classes, attributes with

different values, relationships between types and

properties). Usually, users do not want to see these

types simultaneously, due to the cognitive overload

it would arise.

For example, clicking on an X-class, relations

with classes Y and Z are displayed. These classes

could be highlighted, while other parts of the

ontology could loose focus. The highlight could be

obtained through visual attributes such as colour,

transparency, shapes and positioning. This feature

would be very useful to get an idea of organizing a

mereology (part-to-whole relationship, part-to-part

relationship). Clicking on a main class could be easy

to identify the classes that represent the parts.

The relationships properties (transitivity,

reflexivity, symmetry, if it is functional or not) are

an important structural component, because they

have impact onto the inference that can be

performed with the ontology. Likewise, the

attributes of each class (data properties) should be

considered in the visualization.

Regarding question (3), the main problems of

current tools for ontologies visualization are

common to any tool for graph visualization:

problems of scale versus amount of information that

needs to be presented. An alternative would be to

use different visualization techniques.

According to Gurr (1999), visual representations

can be constructed in order to express the properties

of a concept. The use of tooltip texts can help in the

encoding of the displayed information, because they

contain high loads of information and are presented

selectively as the user explores the visualization of

the ontology.

Finally, a simple but important suggestion from

the users was that views of ontologies fit on an A4

format, with sufficient level of detail. It would also

be interesting to have a tool that allows adding and

removing elements of the visualization in a quick

and simplified mode.

3.2 Requirements

From such results, we can list the following

requirements for ontology visualization:

 Provide overview of hierarchy ontology, with the

possibility of detailing some parts;

 Avoid presenting the different aspects of an on-

tology (classes, description, object properties, data

properties, individuals) together in a unique

visualization;

 Optimize the results from the ontology

validation;

 Explore the use of visual attributes such as

colour, transparency, and shapes;

 Provide display filters based on different

techniques of focus+context and/or overview+detail,

zoom, pan and rotation of the image;

 Allow rapid and simple inclusion of visual

elements in the visualization, as well as their

removal;

 Allow printing the entire ontology in paper sizes

commonly used, such as A4.

Considering these aspects, in the next section, we

present the initial idea of our method to assist the

user in the visualization of ontology hierarchy and

relationships.

4 PROPOSED VISUALIZATION

When we analyze an image, we activate our

perceptual mechanisms to identify patterns and

perform segmentation of elements. The user must

perceive the information presented in the display,

and the understanding involves cognitive

processes. An image can be ambiguous due to lack

of relevant information or by excess of irrelevant

information.

Graphs are the most intuitive form of visualizing

the relationships between concepts of ontologies by

their both hierarchical and relational

characteristics. However, relationships are displayed

in expanded nodes and the overlapping edges can be

a problem for the efficiency of information

display. An interactive graph or tree solves part of

the problem, allowing the user to highlight the

information in focus through selection, but the

overlapping edges are still a problem. In this sense,

Katifiori (2007) list tasks related to ontologies and

visualizations as shown in Table 1.

We studied the hypothesis of representing

ontologies in a 3D space, allowing the user to

navigate through in-depth visual representations,

rotating, expanding and selecting the desired

items. However, such views require the user

immersion and depth perception is crucial.

Considering these aspects, we propose a

visualization method that fits the requirements

pointed out by users as well as the tasks listed in

REQUIREMENTS FOR INTERACTIVE ONTOLOGY VISUALIZATION

- Using Hypertree +2.5D Visualization for Exploring Relationships between Concepts

245

Table 1. In the first part of this study, we have

chosen to focus on visualizing the hierarchy of the

ontology and the relationships between concepts. As

an approach to that, the hyperbolic tree is an

overview+detail method, which reduces the

cognitive overload and the user disorientation that

might happen during the interaction with the nodes,

expanding and contracting them, especially in

ontologies with many concepts.

However, besides the class hierarchy

(relationship "is one"), users of ontologies need to

analyse, in an integrated way, the other ontology

relationships. Thus, we actually have a graph along

with the tree, but end up with the problem of

occlusion of information by the overlapping

edges. This problem can be solved with the use of a

third dimension to display one or more relationships

(object properties) selected by the user. To view

them, we take the plane where the tree is displayed

and perform a 90° rotation around the X-axis (see

Figure 2a). The rotated plane, positioned in 3D as an

XZ-plane, displays the hyperbolic tree, and selected

relationships are represented as curved lines in

space, connecting the related concepts (Figure 2b),

without interfering with the display of the

hierarchical relation.

Table 1: Tasks related to ontologies and information

visualization techniques (adapted from Katifiori, 2007).

Task Description VI Techniques

Overview

Gain an overview of

the entire collection.

Trees and graphs,

3D landscapes,

treemaps (space

filling)

Zoom

Zoom in on items of

interest.

Indented lists, trees

and graphs, 3D

landscapes

Details-on-

demand

Select an item or

group and get details

when needed.

Trees and graphs,

3D landscapes,

Filter

Filter out

uninteresting items.

Indented lists, trees

and graphs

Relate

View relationships

among items.

Indented lists, trees

and graphs,

zooming, 3D

landscapes

History

Keep a history of

actions to support

undo, replay and

progressive

refinement.

In addition to rotations around the X-axis, rotations

around the axes Y and Z, zoom and pan are also

allowed, providing full 3D navigation. These

interactions are performed with a control, common

in tools for visualization of georeferenced data and

following usability heuristics "consistency and

standards", Thus, this control becomes more

intuitive for the user in order to facilitate the setting

with the tool.

Figure 2b shows the proposed 2.5D scheme

applied to an ontology hierarchy/graph. We explored

colour and thickness of edges and line contours in

nodes. The user remains "aware" of the ontology

hierarchy and visualizes one or more relationships in

a separate spatial dimension.

Our 2.5D visualization was presented to the four

users after they were interviewed. Informally they

approved the new possibilities for displaying and

interacting with the ontologies represented in that

way.

Also as a preliminary validation, we invited a

fifth specialist, belonging to a group that studies

quality of information on the web and

recommendation systems. She was asked to perform

a new analysis based on three questions: (1) The

initial layout is clear? (2) It is possible to clearly

separate the hierarchy concepts of the relationships

between theses? (3) The technique is useful for the

exploration of ontology aspects? Three possibilities

of answers were defined: Yes; Partially; No. The

user explored the visualization in many ways,

marking the option “Yes” for all questions.

For sure, we need to perform further studies to

find alternatives to display tooltips related to nodes,

attributes, instances and semantic concepts. Icons,

symbols and transparency are being studied in

addition to other information visualization

techniques.

5 FINAL COMMENTS

Information visualization techniques amplify

cognition and reduce exploration time of a data set,

allowing the recognition of patterns and facilitating

inferences about different concepts. We have

designed a visual and interactive way to explore

ontologies, improving the process of insight from

such data.

In this work, we also discussed requirements for

visualization and interaction with ontologies in order

support our approach to help users to perform more

easily and efficiently the different operations on the

ontology. Considering these aspects, we have

proposed a 2.5D visualization of ontologies that

combine aspects of both 2D and 3D views and take

into account the aspects pointed out by the expert

users. The main idea is to provide a representation

that is intuitive and allows efficient analysis of the

IVAPP 2011 - International Conference on Information Visualization Theory and Applications

246

Figure 2: (a) Rotation of the plane with ontology hierarchy and relationships; (b) 2.5D ontology visualization.

concepts displayed in the ontology visualization.

This study represents an initial step in the

development of an ontology visualization

tool. Future work involves the investigation of

alternative display of properties and instances of an

ontology in addition to other requirements listed in

Section 3.

ACKNOWLEDGEMENTS

We would like to thank the users that participated in the

interviews and further evaluation process.

REFERENCES

Amaral, F. Visualizing the semantics (not the syntax) of

concept descriptions. In VI Workshop em Tecnologia

da Informação e da Linguagem Humana (TIL 2008),

Vila Velha, ES, 2008.

Babaria, K. Using Treemaps to Visualize Gene Ontologies,

Human Computer Interaction Lab and Institute for

Systems Research, University of Maryland, College

Park, MD USA, 2004.

Baehrecke, E. H., Dang, N., Babaria, K. Shneiderman, B.

Visualization and analysis of microarray and gene

ontology data with treemaps. BMC Bioinformatics.

2004.

Bosca, A., Bomino, D., Pellegrino, P. OntoSphere: more

than a 3D ontology visualization tool. In Proceedings

of SWAP, the 2nd Italian Semantic Web Workshop,

Trento, Italy, December 14-16, CEUR, Workshop

Proceedings, ISSN 1613-0073, Vol-166, 2005.

Catenazzi, N., Sommaruga, L., Mazza, R. User-friendly

ontology editing and visualization tools: the

OWLeasyViz approach. In: Proceedings of the 13th

IEEE International Conference on Information

Visualisation. Barcellona, Spain. 14-17 July 2009. pp.

283-288. IEEE. ISBN: 978-0-7695-3733-7.

Fluit, C., Sabou, M., Harmelen, F. Ontology-based

Information Visualisation: Towards Semantic Web

Applications. International Symposium of

Visualisation of the Semantic Web (VSW'05). 2005.

Freitas, C. Visualização de Informações e a Convergência

de Técnicas de Computação Gráfica e Interação

Humano-Computador. Jornadas de Atualização em

Informática (JAI), XXVII Congresso da SBC, 2007.

Gruber, T. (1996). What is an ontology? [S.l.: s.n.], 1996.

DOI=http://www-ksl.stanford.edu/ kst/ what-is-an-

ontology.html. Último acesso em: abril de 2009.

Gurr, C. Effective Diagrammatic Communication:

Syntatic, Semantic and Pragmatic Issues, Journal of

Visual Languages and Computing, 10, 317-342, 1999.

Katifori, A.; Halatsis, C.; Lepouras, G.; Vassilakis, C.;

Giannopoulou, e. Ontology Visualization Methods - A

Survey. ACM Comput. Surv. 39, 4 (Nov. 2007), 10.

Katifori A, Torou E, Vassilakis C, Lepouras G, Halatsis

C: Selected results of a comparative study of four

ontology visualization methods for information

retrieval tasks. In: Research Challenges in Information

Science, 2008 RCIS 2008 Second International

Conference on: 2008; 2008: 133-140.

Kriglstein, S. User Requirements Analysis on Ontology

Visualization. Proc. International Conference on

Complex, Intelligent and Software Intensive Systems.

2nd International Workshop on Ontology Alignment

and Visualization, Fukuoka, Japan, 2009.

Kriglstein, S. Wallner, G. Knoocks - A Visualization

Approach for OWL Lite Ontologies. CISIS 2010, The

Fourth International Conference on Complex,

REQUIREMENTS FOR INTERACTIVE ONTOLOGY VISUALIZATION

- Using Hypertree +2.5D Visualization for Exploring Relationships between Concepts

247

Intelligent and Software Intensive Systems, Krakow,

Poland, 15-18 February 2010.

Lanzenberger, M., Sampson, J., Rester, M. Visualization

in Ontology Tools. Proc. International Conference on

Complex, Intelligent and Software Intensive Systems.

2nd International Workshop on Ontology Alignment

and Visualization, Fukuoka, Japan, 2009.

Nielsen, J. Usability Inspection Methods. Proceedings of

Conference on Human Factors in Computing Systems

(CHI’95). 1994. Colorado, USA.

Noy, N., Fergerson, R., Musen, M. The knowledge model

of Protege-2000: Combining interoperability and

flexibility. In Proceedings of 2nd International

Conference on Knowledge Engineering and

Knowledge Management (EKAW'2000), Juanles-Pins,

France, 2000.

Noy, N.; McGuiness, D. Ontology Development 101 – A

guide to creating your first ontology. KSL Technical

Report, Standford University, 2001.

Rocha, H., Baranauskas, M. Design e Avaliação de

Interfaces Humano-Computador. Campinas,

SP:NIED/UNICAMP, 2003.

Samper, J., Tomás, V., Carrillo, E., Nascimento, R.

Visualization of ontologies to specify semantic

descriptions of services. IEEE Transactions on

Knowledge and Data Engineering. 20(1): p. 130-134.

2008.

Schevers, H. A. J., Trinidad, G.; Drogemuller, R.M.

Towards Integrated Assessments for Urban

Development. Journal of Information Technology in

Construction (ITcon), Vol. 11, Special Issue Decision

Support Systems for Infrastructure Management, pg.

225-236. DOI=http://www.itcon.org/2006/17. Último

acesso em: outubro de 2010.

Sowa, J. F. Guided Tour of Ontology, 2005.

DOI=http://www.jfsowa.com/ontology/guided.htm.

Silva, I., Freitas, C. Avaliação de Ferramentas de Busca na

Web baseadas em Visualização de Informações. In:

Simpósio de Fatores Humanos em Sistemas

Computacionais (IHC), 2006, Natal - RN.

IVAPP 2011 - International Conference on Information Visualization Theory and Applications

248