ArticlePDF Available

Visualisation of learning ontologies

Authors:
Trent Apted
Trent Apted
Trent Apted
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Judy Kay at The University of Sydney
Judy Kay
Andrew Lum at The University of Sydney
Andrew Lum
James Uther at The University of Sydney
James Uther
Download full-text PDF
Read full-text
Download citation

Abstract

This paper describes a novel system designed to build scrutable student models from minimal information about the student. It uses MECURIO, our tool for automatic generation of ontologies from online sources. The system takes a simple statement of the student's knowledge or interest in a topic. It uses the ontology to generate a larger set of components for the student model. The student is then able to scrutinise these inferences with a visualisation tool, VlUM.
Content uploaded by Judy Kay
Author content
All content in this area was uploaded by Judy Kay on Feb 09, 2016
Content may be subject to copyright.
Visualisation of learning ontologies
Trent APTED, Judy KAY, Andrew LUM, James UTHER
School of Information Technologies,
University of Sydney, AUSTRALIA 2006
Abstract. This paper describes a novel system designed to build scrutable student
models from minimal information about the student. It uses MECURIO, our tool for
automatic generation of ontologies from online sources. The system takes a simple
statement of the student’s knowledge or interest in a topic. It uses the ontology to
generate a larger set of components for the student model. The student is then able to
scrutinise these inferences with a visualisation tool, VlUM.
1. Introduction
Ontologies will play an important role in many aspects of future teaching systems,
including such diverse tasks as supporting searches for relevant learning objects, generating
teaching elements and supporting reasoning about the learner's knowledge. It is the last of
these that is the particular focus of our work.
Although ontologies have the potential to be very useful, they are generally difficult
and time-consuming to construct manually. We have been exploring ways to automatically
construct a light-weight ontology [1] from existing documents and, in particular, from
existing dictionaries. This has been used in a tool for visualising large student models. The
ontology provides the structure that enables the visualisation tool to operate effectively.
The user can select a focus concept on the display and the ontology is then used to ensure
that the most closely related concepts are visible. This selection of concepts to be made
visible is an essential part of the visualisation that assists learners in exploring domains
with hundreds of concepts. The particular task we want to support involves showing a user
what an ontology allows us to infer from information such as the learner’s own assessment
of their knowledge.
2. Automated construction of a large ontology
We have used MECUREO [2] to build the ontologies that we use in this project. It
parses a resource such as FOLDOC, the Free On-Line Dictionary Of Computing [3]. The
process results in an ontology of computer science. This is represented as a weighted
digraph. Each node in the ontology is a concept from the dictionary. Links between these
have weights (0, 1]. These associate a cost with the link, with a smaller weight or cost
meaning that the concepts are more strongly related. Links also have a type. Types have an
associated direction. MECUREO builds a large ontology from FOLDOC: 23,095 concepts
and 57,550 directed relationships, with 55,038 of these between distinct concepts.
Essentially, we use point queries to grow student models. So, for example, if a learner
claims to know the concept Haskell, we make a point query on this term, with a suitable
cost-distance value such as 2.0. This returns a subgraph within the full ontology. We use
this to make additional inferences about the learner’s likely knowledge. Similarly, if a
learner expresses interest in learning about Haskell, we can use the same approach to build
a model of a wider set of likely interests. This approach has the potential to help address the
problems of the narrow bandwidth of communication between the user and the machine.
We now describe how we use this ontology to support visualisation of a user model.
3. Ontology visualisation
VlUM, for Visualisation of Large User Models, can effectively display large user models in
web-based systems [4].
VlUM displays the components of the learner model in a vertical listing. It utilises
perspective distortion to enable users to navigate the user model. At any point in time, the
component with the largest font is the one currently selected. Figure 1 shows three VlUM
screens, each displaying an ontology generated from FOLDOC with the concept Microsoft
Word currently selected.
Topics closely related to the selected component appear in a larger font size, spacing
and brightness than those further away in the underlying graph. We can see related
concepts such as GUI, and MS-DOS clearly. Concepts that are not relevant are bunched
together in a small dimmed font.
Figure 1: The visualisation for a MECUREO generated ontology grown from a point query of the concept
Microsoft Word (currently selected) with depth slider set at 1, 2 and 3 from left to right. The depth slider
controls the number links out from the focus concept. The greater the value, the more concepts appear on the
display.
Users can navigate through the model by clicking on a concept to select it. The
display changes so that the newly selected concept becomes the focus. It now appears as
the largest word and a new spanning tree from that concept is generated.
4. Evaluation
We designed MECUREO so that we could take a small amount of information about a user,
since it is common that this is all that is practically available, and use this information to infer
many more things about the user. The role of VlUM is to make that new user model available
to the user so that they can scrutinise it. In particular, we would like users to be able to check
that they are happy with the inferences that have been made.
We have performed a qualitative experiment designed to determine whether users could
indeed make effective use of this display. The participants in the evaluation were five upper
high school students attending a summer school in which they would learn to program in
Python, and five undergraduate computer science students. We asked them to perform tasks on
an ontology from a point query on Python. These users are representative of a quite computer
literate group. Overall, the experiment indicated that the participants could readily
appreciate the general principle of the system, taking a single piece of user input and using
that to generate a much larger user model. All appeared to be able to use it effectively for
the experiment.
5. Conclusions
There are several innovative aspects to the work we have described. First, there has been
very little work on building ontologies automatically from existing resources such as
disctionaries and much that has been done has used much heavier weight approaches, as for
example Kietz et al [5]. There is also the challenging task of evaluating ontologies. A third
contribution of this work is its exploration of visualisation of large user models. There has
been some work on visualisation of a single node and its neighbours in a student model by
Zapata-Rivera and Greer [6]. The combination of all these elements is a further innovation.
The qualitative evaluation we have described is rather limited. It was also designed
to be easily understood by users in a short period of time. So, it does not directly assess
issues of student learning. The importance of our work, for learning, follows from the range
of research on supporting learning through reflection on the student model, such as
explored by Dimitrova et al [7], Bull [8] and Self [9]. Even so, our evaluation indicates the
promise of the approach. It also indicates that the participants in our study could make
sense of the task and could use the VlUM display to explore the ontology. They also shared
our appreciation of the power of this approach to building detailed learner models from
minimal user input.
References
1. Mizoguchi, R., Ontology-based systemization of functional knowledge. 2001.
2. Apted, T. and Kay, J. Automatic Construction of Learning Ontologies. In: L. Aroyo and D. Dicheva,
Editors. International Conference on Computers in Education; 2002. Technische Universiteit
Eindhoven, p. 55-62.
3. FOLDOC: The Free Online Dictionary of Computing, 2001, Available at http://www.foldoc.org/.
4. Uther, J., On the Visualisation of Large User Model in Web Based Systems, PhD Thesis, 2001,
University of Sydney.
5. Kietz, J.-U., Volz, R., and Maedche, A. Extracting a Domain-Specific Ontology from a Corporate
Intranet. In: C. Cardie, et al., Editors. Computational Natural Language Learning and of the Second
Learning Language in Logic Workshop; 2000, p. 167-175.
6. Zapata-Rivera, J.D. and Greer, J., Inspecting and Visualizing Distributed Bayesian Student Models, in
G. Gauthier, C. Frasson, and K. VanLehn, Editors, Intelligent Tutoring Systems. 2000. p. 544-553.
7. Dimitrova, V., Self, J.A., and Brna, P. The interactive maintenance of open learner models. In:
International Conference on Artificial Intelligence in Education; 1999, p. 405-412.
8. Bull, S., Brna, P., and Pain, H., Extending the scope of the student model. User Modeling and User-
Adapted Interaction, 1995. 5(1): p. 44-65.
9. Self, J.A., Open sesame?: fifteen variations on the theme of openness in learning environments.
International Journal of Artificial Intelligence in Education, 1999. 10: p. 1020-1029.
Acknowledgements
We would like to thank Hewlett-Packard for funding parts of this research.
... Several projects, originated in the field of user modelling, apply ontology learning techniques to learn the structure of the domain ontology and then use it for adaptation purposes. For example, the ontology learnt by MECUREO ) (which we mentioned in Section 3.4.2) has been applied in a number of adaptive systems (for example, for visualisation and navigation (Apted et al., 2003a) and for learning object metadata generation ). Zouaq et al. (2007) reports initial progress on the Knowledge Puzzle system that tries to learn domain ontologies from unstructured text. ...
... According to the student evaluation of several layouts for opening conceptual knowledge models, concept maps and the lecture structure have been found most useful. The VlUM project (Uther and Kay, 2003;Apted et al., 2003aApted et al., , 2003b proposes an interesting solution to the problem of visualisation of large ontology-based user models. When the size of the domain ontology is above several hundred concepts, visualising the user model can become a challenging task, as the capabilities of the screen as well as the perception capabilities of the users are limited. ...
Ontological technologies for user modelling
Article
Full-text available
  • Apr 2010
This paper brings together research from two different fields – user modelling and web ontologies – in attempt to demonstrate how recent semantic trends in web development can be combined with the modern technologies of user modelling. Over the last several years, a number of user-adaptive systems have been exploiting ontologies for the purposes of semantics representation, automatic knowledge acquisition, domain and user model visualisation and creation of interoperable and reusable architectural solutions. Before discussing these projects, we first overview the underlying user modelling and ontological technologies. As an example of the project employing ontology-based user modelling, we present an experiment design for translation of overlay student models for relative domains by means of ontology mapping.
View
... -Dans le projet de repérage et de visualisation du model de l'apprenant de (Apted T., Kay J., Lum A. et Uther J., 2003), un outil de repérage et de visualisation, VIUM, a été conçu pour permet à l'usager de sélectionner un concept central sur l'écran. Une ontologie est utilisée pour s'assurer que les concepts les plus proches sémantiquement sont visibles. ...
Apport de l'ingénierie ontologique aux environnements de formation à distance
Preprint
  • Jan 2003
Cet article explore le potentiel de l'ontologie et de l'ingénierie ontologique pour augmenter l'intelligence dans les environnements de formation à distance, ainsi que dans la conception de ces environnements. Il contient une introduction à l'ontologie avec ses origines dans la philosophie occidentale, une discussion de s relations entre l'intelligence artificielle et l'ingénierie ontologique, suivie d'une description des méthodologies et des outils d'ingénierie ontologique. Plusieurs pistes d'exploitation du potentiel des ontologies pour les environnements de formation à distance sont explorées. ABSTRACT. How is Ontology relevant to Distance Education? This article claims that ontology and ontological engineering have a potential to increase intelligence both in the learning environments for distance education, and in the process of designing them. An introduction to the origins of the concept of ontology in western philosophy is presented, as well as a discussion on the relationships between artificial intelligence and ontological engineering, followed by a description of methodologies and tools for ontological engineering. MOTS-CLES. ontologie, ingénierie ontologique, formation à distance,
View
... Once found, the mapping can be used as a template for the translation of the targeted overlay student models. Several projects use ontologies for representing domain models [Apted et al., 2003; Henze et al., 2004], student profiles [Dolog & Nejdl, 2003; Ounnas et al., 2006] and other knowledge components of AES [Dolog & Nejdl, 2007; Mizoguchi et al., 1996]. Some attempts have been made to implement ontology-based interoperable student models. ...
Ontology-Based Open-Corpus Personalization for E-Learning
Article
  • Jan 2008
Conventional closed-corpus adaptive information systems control limited sets of documents in predefined domains and cannot provide access to the external content. Such restrictions contradict the requirements of today, when most of the information systems are implemented in the open document space of the World Wide Web and are expected to operate on the open-corpus content. In order to provide personalized access to open-corpus documents, an adaptive system should be able to maintain modeling of new documents in terms of domain knowledge automatically and dynamically. This dissertation explores the problem of open-corpus personalization and semantic modeling of open-corpus content in the context of e-Learning. Information on the World Wide Web is not without structure. Many collections of online instructional material (tutorials, electronic books, digital libraries, etc.) have been provided with implicit knowledge models encoded in form of tables of content, indexes, headers of chapters, links between pages, and different styles of text fragments. The main dissertation approach tries to leverage this layer of hidden semantics by extracting and representing it as coarse-grained models of content collections. A central domain ontology is used to maintain overlay modeling of students’ knowledge and serves as a reference point for multiple collections of external instructional material. In order to establish the link between the ontology and the open-corpus content models a special ontology mapping algorithm has been developed. The proposed approach has been applied in the Ontology-based Open-corpus Personalization Service that recommends and adaptively annotates online reading material. The domain of Java programming has been chosen for the proof-of-concept implementation. A controlled experiment has been organized to evaluate the developed adaptive system and the proposed approach overall. The results of the evaluation have demonstrated several significant learning effects of the implemented open-corpus personalization. The analysis of log-based data has also shown that the open-corpus version of the system is capable of providing personalization of similar quality to the close-corpus one. Such results indicate that the proposed approach successfully supports open-corpus personalization for e-Learning. Further research is required to verify if the approach remains effective in other subject domains and with other types of instructional content.
View
... -Dans le projet de repérage et de visualisation du model de l'apprenant de (Apted T., Kay J., Lum A. et Uther J., 2003), un outil de repérage et de visualisation, VIUM, a été conçu pour permet à l'usager de sélectionner un concept central sur l'écran. Une ontologie est utilisée pour s'assurer que les concepts les plus proches sémantiquement sont visibles. ...
Apport de l'ingénierie ontologique aux environnements de formation à distance Contribution of Ontological Engineering to Distance Learning Environments
Article
Full-text available
  • Oct 2003
How is Ontology relevant to Distance Education? This article claims that ontology and ontological engineering have a potential to increase intelligence both in the learning environments for distance education, and in the process of designing them. An introduction to the origins of the concept of ontology in western philosophy is presented, as well as a discussion on the relationships between artificial intelligence and ontological engineering, followed by a description of methodologies and tools for ontological engineering.
View
... This component is used by the user modelling system to decide whether or not to initiate a new user model for that user. In earlier work [1], we have modelled users by exploring evidence of the duration of time the user has spent viewing or interacting with a particular digital object. In MyMuseum, this will map to the time spent at a museum exhibit. ...
Adaptively recommending museum tours
Article
Full-text available
  • Jan 2005
We believe that a user should be able to scrutinise the be-liefs a system has about him/her, namely the user model, as well as be able to update that model. A MyMuseum proto-type of a scrutable and personalised museum guide has been built to explore ways to visualise how the system adapts to the user and the reasons for the adaptation. As a result of the evaluation, we propose to improve the process of rec-ommending a touring path by using an automatically built dynamic ontology.
View
... In order to do so, the tutor has to simply consider the node corresponding to that topic as the root of the domain model and the overlay student model for that session. This is similar to making a point query on a term in a digraph representing an ontology map of the domain in order to extract the relevant student model [1] -by focusing on only the relevant concepts in the student model, the efficiency of accessing and communicating the student model can be significantly improved. We may want to use different criteria of proficiency for different concepts in the domain. ...
Using Enhanced Concept Map for Student Modeling in Programming Tutors.
Conference Paper
Full-text available
  • Jan 2006
We have been using the concept map of the domain. enhanced with pedagogical concepts called learning objectives, as the overlay student model in our intelligent tutors for programming The resulting student model is finegrained, and has several advantages: (I) it facilitates hetter adaptation during problem generation; (2) it makes it possible fur the tutor to automatically vary the level of locality during problem generation to meet the needs of the learner; (3) it clarities to the learner the relationship among domain concepts when opened to scrutiny; (4) the tutor can estimate the level of understanding of a student in any higher-level concept, not just t lie concepts for which it presents problems; and (5) two tutors in a domain can affect each other's adaptation of problems, Prior evaluations have shown that tutors that use enhanced concept maps help improve learning. Copyright © 2006, American Association Artificial Intelligence (www.aaai.ogg). All rights reserved.
View
... -Dans le projet de repérage et de visualisation du model de l'apprenant de (Apted T., Kay J., Lum A. et Uther J., 2003), un outil de repérage et de visualisation, VIUM, a été conçu pour permet à l'usager de sélectionner un concept central sur l'écran. Une ontologie est utilisée pour s'assurer que les concepts les plus proches sémantiquement sont visibles. ...
Apport de l’ingénierie ontologique aux environnements de formation à distance
Article
Full-text available
  • Jan 2003
Cet article explore le potentiel de l’ontologie et de l’ingénierie ontologique pour augmenter l’intelligence dans les environnements de formation à distance, ainsi que dans la conception de ces environnements. Il contient une introduction à l’ontologie avec ses origines dans la philosophie occidentale, une discussion des relations entre l’intelligence artificielle et l’ingénierie ontologique, suivie d’une description des méthodologies et des outils d’ingénierie ontologique. Plusieurs pistes d’exploitation du potentiel des ontologies pour les environnements de formation à distance sont explorées. - English Abstract - How is Ontology relevant to Distance Education? This article claims that ontology and ontological engineering have a potential to increase intelligence both in the learning environments for distance education, and in the process of designing them. An introduction to the origins of the concept of ontology in western philosophy is presented, as well as a discussion on the relationships between artificial intelligence and ontological engineering, followed by a description of methodologies and tools for ontological engineering.
View
... -Dans le projet de repérage et de visualisation du model de l'apprenant de (Apted T., Kay J., Lum A. et Uther J., 2003), un outil de repérage et de visualisation, VIUM, a été conçu pour permet à l'usager de sélectionner un concept central sur l'écran. Une ontologie est utilisée pour s'assurer que les concepts les plus proches sémantiquement sont visibles. ...
Contribution of Ontological Engineering to Distance Learning Environments
Article
Full-text available
  • Jan 2003
How is Ontology relevant to Distance Education? This article claims that ontology and ontological engineering have a potential to increase intelligence both in the learning environments for distance education, and in the process of designing them. An introduction to the origins of the concept of ontology in western philosophy is presented, as well as a discussion on the relationships between artificial intelligence and ontological engineering, followed by a description of methodologies and tools for ontological engineering.
View
Ontology-based systematization of functional knowledge
Article
Full-text available
  • Aug 2004
It has been recognized that design knowledge is scattered around technology and target domains. One of the two major reasons for it is that different frameworks (viewpoints) for conceptualization of design knowledge are used when people t ry to describe knowledge in different domains. The other is that several key functional concepts are left undefined or even unidentified. In this paper, I first overview the state of the art of ontological engineering which I believe is able to make a considerable contribution to resolving these difficulties. I then discuss our enterprise aiming at systematization of functional knowledge used for synthesis. We discuss ontologies that guide conceptualization of artifacts from the functional point of view. The framework for knowledge systematization is based on an extended device ontology and a functional concept ontology built on top of the device ontology. The utilization of the systematized functional knowledge in several application systems is also discussed together with its advantages.
View
Inspecting and Visualizing Distributed Bayesian Student Models
Conference Paper
Full-text available
  • Jun 2000
  • Lect Notes Comput Sci
Bayesian Belief Networks provide a principled, mathematically sound, and logically rational mechanism to represent student models. The belief net backbone structure proposed by Reye (14,15) offers a practical way to rep- resent and update Bayesian student models describing both cognitive and social aspects of the learner. Considering students as active participants in the model- ling process, this paper explores visualization and inspectability issues of Baye- sian student modelling. This paper also presents ViSMod (Visualization of Bayesian Student Models) an integrated tool to visualize and inspect distributed Bayesian student models.
View
Extending the Scope of the Student Model.
Article
Full-text available
  • Mar 1995
In this paper we maintain that there are benefits to extending the scope of student models to include additional information as part of the explicit student model. We illustrate our argument by describing a student model which focuses on 1. performance in the domain; 2. acquisition order of the target knowledge; 3. analogy; 4. learning strategies; 5. awareness and reflection. The first four of these issues are explicitly represented in the student model. Awareness and reflection should occur as the student model is transparent; it is used to promote learner reflection by encouraging the learner to view, and even negotiate changes to the model. Although the architecture is transferable across domains, each instantiation of the student model will necessarily be domain specific due to the importance of factors such as the relevant background knowledge for analogy, and typical progress through the target material. As an example of this approach we describe the student model of an intelligent computer assisted language learning system which was based on research findings on the above five topics in the field of second language acquisition. Throughout we address the issue of the generality of this model, with particular reference to the possibility of a similar architecture reflecting comparable issues in the domain of learning about electrical circuits.
View
Extracting a Domain-Specific Ontology from a Corporate Intranet
Article
Full-text available
  • Dec 2002
This paper describes our actual and ongoing work in supporting semi-automatic ontology acquisition from a corporate intranet of an insurance company. A comprehensive architecture and a system for semi-automatic ontology acquisition supports processing semi-structured information (e.g. contained in dictionaries) and natural language documents and including existing core ontologies (e.g. GermaNet, WordNet). We present a method for acquir- ing a application-tailored domain ontology from given heterogeneous intranet sources.
View
The Interactive Maintenance of Open Learner Models
Article
Full-text available
  • May 2000
. Opening the learner model is a process involving the learner as a collaborator in building a model of his beliefs. The interaction plays a crucial role here since it provides both the system and the learner with a medium to reflect on the learner's beliefs. In this paper we describe a computational framework for the interactive maintenance of open learner models. It adopts some approaches from human-human interaction and considers the two main issues of the interaction -- the language for communication and the underlying dialogue model. We describe a communication language that graphically externalises the learner's beliefs. An approach based on dialogue games is adapted to manage the moves in an interactive diagnostic dialogue. The framework is exemplified in STyLE-OLM, an open learner modelling component in a scientific terminology learning environment. 1 Introduction In traditional student diagnosis, the system seeks to infer the reasons for the learner's behaviour wit...
View
View
View
Automatic construction of learning ontologies
Conference Paper
  • Jan 2003
This paper describes a system which automatically constructs an extensive ontology of computer science. This can serve as a basis for making inferences about student models and other reasoning within a teaching system. The system enables a user to select a focus concept and then see the most closely related concepts.
View
User Modeling and User-Adapted Interaction
  • Jan 1995
  • 44-65
  • S Bull
  • P Brna
  • H Pain
Bull, S., Brna, P., and Pain, H., Extending the scope of the student model. User Modeling and User-Adapted Interaction, 1995. 5(1): p. 44-65.
Open sesame?: fifteen variations on the theme of openness in learning environments
  • Jan 1999
  • 1020-1029
  • J A Self
Self, J.A., Open sesame?: fifteen variations on the theme of openness in learning environments. International Journal of Artificial Intelligence in Education, 1999. 10: p. 1020-1029.