Adaptive Hypermedia
for Education and Training
Peter Brusilovsky
School of Information Sciences, University of Pittsburgh
Pittsburgh, PA 15260, USA
peterb@pitt.edu
SUMMARY
Adaptive hypermedia is a relatively new direction in research at the crossroads of
hypermedia and user modeling. Adaptive hypermedia systems build a model of the goals,
preferences and knowledge of each individual user and use this model throughout the
interaction with the user, in order to adapt to the needs of that user. Educational hypermedia
was one of the first application areas for adaptive hypermedia and is currently one of the
most popular and well-investigated. The goal of this presentation is to explain the nature and
the mechanism of adaptation in educational adaptive hypermedia and to provide several
examples of using adaptive hypermedia in educational and training applications of different
natures and complexity.
KEYWORDS: Adaptive Hypermedia, Web-based Education, Intelligent Tutoring System,
E-learning, Training, Student Model, Personalization
INTRODUCTION
Adaptive hypermedia (AH) is an alternative to the traditional “one-sizefits-all” approach in the development of hypermedia systems. Adaptive
hypermedia (AH) systems build a model of the goals, preferences and
knowledge of each individual user; this model is used throughout the
interaction with the user in order to adapt to the needs of that particular user
(Brusilovsky, 1996b). For example, a student in an adaptive educational
hypermedia system will be given a presentation that is adapted specifically
to his or her knowledge of the subject (De Bra & Calvi, 1998; Hothi, Hall &
Sly, 2000) as well as a suggested set of the most relevant links to proceed
further (Brusilovsky, Eklund & Schwarz, 1998; Kavcic, 2004). An adaptive
electronic encyclopedia will personalize the content of an article to augment
the user's existing knowledge and interests (Bontcheva & Wilks, 2005;
Milosavljevic, 1997). A museum guide will adapt the presentation about
every visited object to the user's individual path through the museum
(Oberlander et al., 1998; Stock et al., 2007).
Adaptive hypermedia belongs to the class of user-adaptive systems
(Schneider-Hufschmidt, Kühme & Malinowski, 1993). A distinctive feature
of an adaptive system is an explicit user model that represents user
knowledge, goals, interests, as well as other features that enable the system
to adapt to different users with their own specific set of goals. An adaptive
system collects data for the user model from various sources that can
include implicitly observing user interaction and explicitly requesting direct
input from the user. The user model is applied to provide an adaptation
effect, i.e., tailor interaction to different users in the same context. In
different kinds of adaptive systems, adaptation effects could vary greatly. In
AH systems, it is limited to three major adaptation technologies — adaptive
content selection, adaptive navigation support, and adaptive presentation.
The first of these three technologies comes from the fields of adaptive
information retrieval (IR) and intelligent tutoring systems (ITS). When the
user searches for information, the system adaptively selects and prioritizes
the most relevant items (Brajnik, Guida & Tasso, 1987; Brusilovsky,
1992b). Adaptive navigation support was introduced in early adaptive
hypermedia systems (de La Passardiere & Dufresne, 1992; Kaplan, Fenwick
& Chen, 1993) and is specifically associated with browsing-based access to
information. When the user navigates from one item to another, the system
can manipulate the links (e.g., hide, sort, annotate) to guide the user
adaptively to the most relevant information items. Adaptive presentation
stems from research on adaptive explanation and adaptive presentation in
intelligent systems (Boyle & Encarnacion, 1994; Paris, 1988). When the
user gets to a particular page, the system can present its content adaptively.
The goal of this paper is to provide an overview of adaptive educational
hypermedia (AEH). The paper, however, neither provides a historicallycentered overview of the field nor offers a detailed classification of AH
technologies, since these reviews can be found elsewhere (Brusilovsky,
2001; Brusilovsky, 2004; Brusilovsky, 2007; Bunt, Carenini & Conati,
2007; Knutov, De Bra & Pechenizkiy, 2009). Instead, the paper attempts to
give a developer-oriented insight into the internal structure of AEH systems.
The remaining part of the paper focuses on three educational hypermedia
design approaches of increasing complexity, illustrating the presentation
with examples from the past research projects of the author. We conclude
with a brief discussion of challenges in the field of adaptive educational
hypermedia.
Adaptive Educational Hypermedia: from Classic
Hypertext to the Adaptive Web
From the very early days of AH, educational hypermedia was one of its
major application areas. In an educational context, users with alternative
learning goals and knowledge of the subjects require essentially different
treatment. In educational hypermedia, the problem of "being lost in
hyperspace" is especially critical. A number of pioneer adaptive educational
hypermedia systems were developed between 1990 and 1996. These
systems can be roughly divided into two research streams. The systems of
one of these streams were created by researchers in the area of intelligent
tutoring systems (ITS), who were trying to extend traditional student
modeling and adaptation approaches developed in this field to ITS with
hypermedia components (Beaumont, 1994; Brusilovsky, 1993; Gonschorek
& Herzog, 1995; Pérez, Gutiérrez & Lopistéguy, 1995). The systems of
another stream were developed by researchers working on educational
hypermedia in an attempt to make their systems adapt to individual students
(De Bra, 1996; de La Passardiere & Dufresne, 1992; Hohl, Böcker &
Gunzenhäuser, 1996).
ADAPTIVE EDUCATIONAL HYPERMEDIA: THE SECOND
GENERATION
Despite the number of creative ideas explored and evaluated in the early
educational AH systems, it was not until 1996 that this research area
attracted attention from a larger community of researchers. This process was
stimulated by the accumulation and consolidation of research experience in
the field. The research in adaptive hypermedia performed and reported on
up to 1996 provided a good foundation for the new generation of research.
While early researchers were generally not aware of each other's work,
many papers published since 1996 were clearly based on earlier research.
These papers cite earlier work, and usually propose an elaboration or an
extension of techniques suggested earlier. In addition, the Web, with its
clear demand for personalization served to boost adaptive hypermedia
research, providing both a challenge and an attractive platform. Almost all
the papers published before 1996 describe classic pre-Web hypertext and
hypermedia. In contrast, the majority of papers published since 1996 are
devoted to Web-based adaptive hypermedia systems.
In the field of educational adaptive hypermedia, the major driving factor
behind second-generation adaptive educational hypermedia was Web-based
education. The imperative to address the needs of the heterogeneous
audience for Web-based courses individually was clear to many researchers
and practitioners. A few early adaptive hypermedia systems developed for
Web-based education context by 1996, such as ELM-ART (Brusilovsky,
Schwarz & Weber, 1996b), InterBook (Brusilovsky, Schwarz & Weber,
1996a), and 2L670 (De Bra, 1996), provided "proof of existence" and
influenced a number of more recent systems. The majority of adaptive
educational hypermedia systems developed since 1996 are Web-based
systems which were developed for Web-based education context. Some
earlier examples are: ADI (Schöch, Specht & Weber, 1998), RATH
(Hockemeyer, Held & Albert, 1998), ACE (Specht & Oppermann, 1998),
TANGOW (Carro, Pulido & Rodríguez, 1999), Arthur (Gilbert & Han,
1999), CAMELEON (Laroussi & Benahmed, 1998), KBS-Hyperbook
(Henze et al., 1999), AHA! (De Bra & Calvi, 1998), and Multibook
(Steinacker et al., 1999).
The choice of the Web as a development platform turned out to be a wise
one for educational hypermedia systems. It extended the life of a number of
pioneer systems. In particular, the first Web-based adaptive educational
hypermedia systems developed before 1996 such as ELM-ART, InterBook,
and 2L670 are still in use and have been significantly updated and extended
to incorporate a number of new techniques were used for several
experimental studies (Brusilovsky & Eklund, 1998; De Bra & Calvi, 1998;
Weber & Brusilovsky, 2001) that further guided development of the field.
The work on second-generation adaptive educational hypermedia was
performed mainly between 1996 and 2002. It can be roughly split into three
different streams which lack clear-cut borders. The largest group of work
(produced mainly by researchers coming from the Web-based education
side) focused on creating adaptive Web-based educational systems with
elements of adaptive hypermedia. The main motivation was to produce
systems to be used in teaching, not in developing new technologies. As a
result, the works of this stream broadly re-used already existing
technologies and explored various subject areas and approaches. A smaller
stream of work (produced mainly by researchers who were very familiar
with ITS or the adaptive hypermedia area) focused on producing new
techniques for adaptive hypermedia. For example the early AHA! project
(De Bra & Calvi, 1998) explored several approaches to link removal.
MetaLinks (Murray et al., 2000) explored advanced approaches to
hyperspace structuring. INSPIRE explored the use of learning styles
(Papanikolaou et al., 2003) and MANIC (Stern & Woolf, 2000) explored
innovative approaches for user modeling and adaptive presentation. Finally,
another stream of work (which was small, but rapidly expanded) focused on
developing frameworks and authoring tools for producing adaptive
hypermedia systems. The majority of this work produce hat we can call
frameworks for adaptive Web-based education: KBS-Hyperbook (Henze et
al., 1999), Multibook (Steinacker et al., 1999), ACE (Specht & Oppermann,
1998), CAMELEON (Laroussi & Benahmed, 1998), MediBook (Steinacker
et al., 2001), and ECSAIWeb (Sanrach & Grandbastien, 2000). While not
resulting in end-user authoring tools, a framework typically introduces a
generic re-usable architecture and approach that could be used to produce a
range of adaptive systems with low overhead. A few of the most
experienced teams, those working on adaptive hypermedia projects for
several years, introduced practical authoring systems that could be utilized
by end-users to develop adaptive hypermedia systems and courses.
Examples are InterBook (Brusilovsky et al., 1998), ART-Web/NetCoach
(Weber, Kuhl & Weibelzahl, 2001), AHA! (De Bra & Calvi, 1998) and
MetaLinks (Murray et al., 2000).
ADAPTIVE EDUCATIONAL HYPERMEDIA: THE THIRD GENERATION
Altogether, the systems of the second-generation adaptive educational
hypermedia demonstrated a variety of ways to integrate adaptation
technologies into Web-based education systems as well as the value of these
technologies. Yet, they failed to influence practical Web-based education.
Almost 10 years after the appearance of the first adaptive Web-based
educational systems, just a handful are used for teaching real courses,
typically for a class led by one of the authors of the adaptive system.
Instead, the absolute majority of Web-enhanced courses rely on so-called
learning management systems (LMS). LMS are powerful integrated systems
that support a number of needs of both teachers and students. Teachers can
use a LMS to develop Web-based course notes and quizzes, to communicate
with students and to monitor their progress. Students can use it for
communication and collaboration. The complete dominance of LMS over
adaptive systems may look surprising. Actually, for every function that a
typical LMS performs, we can find an adaptive Web-based Educational
System (AWBES) that can significantly outperform the LMS. Adaptive
textbooks created with systems like AHA!, InterBook or NetCoach
mentioned above can help students learn faster and better. Adaptive quizzes
delivered by such systems as SIETTE (Conejo, Guzman & Millán, 2004)
and QuizGuide (Hsiao, Sosnovsky & Brusilovsky, 2010) evaluate student
knowledge more precisely with fewer questions. Adaptive class monitoring
systems (Oda, Satoh & Watanabe, 1998) give the teachers more
opportunities to notice students that are lagging behind. Adaptive
collaboration support systems (Soller, 2007) can reinforce the power of
collaborative learning. It seems obvious that the drawback to modern
adaptive systems is not the quality of their performance, but their inability to
meet the needs of practical Web-enhanced education. The challenge of
integrating adaptive hypermedia technologies into the regular educational
process has defined the current third generation of adaptive educational
hypermedia research.
Various research groups stress different reasons for the domination of
LMS and thus, pursue different research directions. One research stream
focused on the versatility of LMS, attempting to provide in one system as
many teacher and learner support features (from content authoring to
quizzes to discussion forums) as provided by a modern LMS -– plus, the
ability to adapt to the user (Morimoto et al., 2007; Specht et al., 2002; Ueno,
2005). A different stream addressed another superior feature of an LMS -the ability to integrate open corpus Web content. The systems in this stream
explored several approaches to integrating open corpus content in an
adaptive hypermedia system while providing adaptive guidance for this
content (Brusilovsky, Chavan & Farzan, 2004; Brusilovsky & Henze, 2007;
Henze & Nejdl, 2001). Most recent projects, however, choose not to
compete with present-day LMS, but instead to focus on adaptive features of
the coming generation of Web-based educational systems. This new
generation, which will replace modern LMS, will be based on system
interoperability and reusability of content and supported by a number of
emerging E-Learning interoperability. A number of research teams are
trying now to integrate existing adaptive hypermedia technologies with the
ideas of standard-based reusability (Conlan, Dagger & Wade, 2002; Dolog
et al., 2003; Morimoto et al., 2007). However, other teams argue that the
current generation of standards is not able to support the needs of adaptive
learning (Mödritscher, García Barrios & Gütl, 2004; Rey-López et al.,
2008). Yet another direction of work attempts to explore the ideas of the
Semantic Web for content representation and resource discovery,
capitalizing on standards such as Resource Description Framework (RDF)
and Topic Maps (Denaux, Dimitrova & Aroyo, 2005; Dichev, Dicheva &
Aroyo, 2004; Dolog et al., 2003; Henze, 2005; Jacquiot, Bourda &
Popineau, 2004)(Dolog & Nejdl, 2007).
Adaptive Educational Hypermedia: A Designer’s View
2.1 KNOWLEDGE BEHIND PAGES
Despite an amazing diversity of existing AEH systems, almost all of them
are based on the same set of design principles. It is important for those who
are interested in applying or developing AEH systems to understand these
principles. The key to intelligence and adaptivity in these systems is the
presence of a knowledge space (formed by topics, concepts, rules or other
kinds of knowledge elements) beyond the traditional hyperspace formed by
interconnected pages (Fig. 1).
The knowledge space (also known as the domain model) serves as the
backbone for AEH systems. It is used to structure the information about
individual user knowledge and goals (known as the user model or student
model in AEH systems). It is also used to describe the content of
information pages in these systems. In this capacity, the knowledge space
empowers a range of specific AH technologies (such as adaptive sequencing
or adaptive link annotation) to bridge the gap between user knowledge and
goals on one side and the information content on the other side. Such
technologies help the user to receive the most appropriate educational or
training content. While the general principles of knowledge structuring and
user modeling are shared by the majority of AES systems, practical system
may differ a great deal in their complexity and the range of supported
adaptation techniques. More specifically, larger and more diverse
information spaces typically require more sophisticated approaches to
information indexing (i.e., connecting information pages with knowledge
elements) and user modeling. For example, systems with a small
information space (such as those developed in the early days of adaptive
hypermedia) frequently use just one concept to describe an information
fragment. Larger information spaces – with many pages related to the same
concept – demand more precise multi-concept indexing to make pages more
distinct from the system’s point of view. In turn, these more sophisticated
approaches enable a wider range of adaptation techniques. Following earlier
reviews (Brusilovsky, 1996a; Brusilovsky, 2003) three groups of
information indexing approaches of increasing complexity are identified.
The analysis of these three groups is the focus of the second part of this
paper. After a brief introduction to the principles of domain modeling and
student modeling in AEH systems, the remaining part of the paper analyzes
these major information indexing approaches one by one, illustrating each
with an detailed practical example.
Figure 1. The key to adaptivity in AEH systems is the knowledge layer behind the traditional
hyperspace
2.2 THE DOMAIN MODEL
The heart of the knowledge-based approach to developing adaptive
hypermedia systems is a structured domain model that is composed of a set
of small domain knowledge elements (KE). Each KE represents an
elementary fragment of knowledge for the given domain. KE can be named
differently in different systems—concepts, knowledge items, topics,
knowledge elements, learning objectives, learning outcomes; however, in all
cases, they denote elementary fragments of domain knowledge. Depending
on the domain, the application area, and the choice of the designer, KE can
represent bigger or smaller pieces of domain knowledge. A set of KE forms
a domain model. More exactly, a set of independent KE is the simplest form
of domain model. It is typically called a set model or a vector model
(Brusilovsky, 2003) since the set of KE has no internal structure. In a more
advanced form of domain model, KE are related to each other thus forming
a semantic network. This network represents the structure of the domain
covered by a hypermedia system. This kind of model is known as a network
model (shown on the left part of Fig. 1).
The structured domain model was inherited by adaptive educational
hypermedia systems from the field of ITS, where it was used mainly by
systems with task sequencing, curriculum sequencing, and instructional
planning functionality (Brecht, McCalla & Greer, 1989; Brusilovsky,
1992a). This model proved to be relatively simple and powerful and was
later accepted as the de-facto standard by almost all educational and many
non-educational adaptive hypermedia systems.
Domain models in AEH systems seriously differ in complexity. Some
systems developed for teaching practical university courses employed only
the simplest vector domain model (Brusilovsky & Anderson, 1998; De Bra,
1996). At the same time, a number of modern AEH systems use
sophisticated ontology-based networked models with several kinds of links
that represent different kinds of relationships between the KE. The most
popular kind of links in AEH are prerequisite links between the KE. A
prerequisite link represents the fact that one of the related KE has to be
learned before another. Prerequisite links are relatively easy to understand
by authors of educational systems and can support several adaptation and
user modeling techniques. In many AEH systems, prerequisite links are the
only kind of links between KE (Davidovic, Warren & Trichina, 2003;
Farrell et al., 2003; Henze & Nejdl, 2001; Papanikolaou et al., 2003). Other
types of links which are popular in many systems are the classic semantic
links, "is-a" and "part-of" (De Bra, Aerts & Rousseau, 2002a; Hoog et al.,
2002; Steinacker et al., 2001; Trella, Conejo & Bueno, 2002; Vassileva,
1998). The popularity of these links is currently increasing following the
expanded use of more formal ontologies in place of domain models
(Dagger, Wade & Conlan, 2004; Mitrovic & Devedzic, 2004; TrausanMatu, Maraschi & Cerri, 2002).
Another difference in complexity is related to the internal structure of
concepts. For the majority of AEH systems, the domain concepts are
nothing more than names that denote fragments of domain knowledge. At
the same time, some AH systems use a more advanced frame-like
knowledge representation; i.e., represent the internal structure of each
concept as a set of attributes or aspects (Beaumont, 1994; Brusilovsky &
Cooper, 2002; Hohl et al., 1996; Weber & Brusilovsky, 2001).
2.3 THE STUDENT MODEL
One of the most important functions of the domain model is to provide a
framework for representation of the user's domain knowledge. The majority
of AEH systems use an overlay model of user knowledge (also known as an
overlay student model). The overlay model was also inherited from the field
of ITS. The key principle of the overlay model is that for each domain KE,
the individual user knowledge model stores some data that is an estimation
of the user’s knowledge level for this KE. In the simplest (and oldest) form,
it is a binary value (known – not known) that enables the model to represent
the user's knowledge as an overlay of domain knowledge. While some
successful AEH systems (De Bra, 1996) use this classic form of an overlay
model, the majority of systems use a weighted overlay model that can
distinguish several levels of the user's knowledge of a KE through a
qualitative value (Brusilovsky & Anderson, 1998; Papanikolaou et al., 2003)
(for example, good-average-poor), an integer numeric value (for example,
from 0 to 100) (Brusilovsky et al., 1998; De Bra & Ruiter, 2001), or a
probability that the user knows the concept (Henze & Nejdl, 1999; Specht &
Klemke, 2001). A few AEH systems use an even more sophisticated layered
overlay model (Brusilovsky & Millán, 2007) to store multiple evidences
about the user’s level of knowledge separately (Brusilovsky & Cooper,
2002; Brusilovsky, Sosnovsky & Yudelson, 2005; Weber & Brusilovsky,
2001). The level of sophistication in student modeling has been constantly
increasing to support increasingly sophisticated personalization needs and
we expect this process will continue in the context of lifelong modeling
(Kay & Kummerfeld, 2010).
All kinds of weighted overlay models are known to be powerful
personalization tools due to their ability to independently assess and store
the evidences of the user's knowledge about different KE. This power can be
further extended by taking into account connections between KE
represented in the domain model and using them for weight propagation
between KE. Weight propagation increases the impact of a single
observation (such as answering a single question) on the student model and
decreases student modeling sparsity. Good examples of student models
incorporating weight propagation are Bayesian student models (Brusilovsky
& Millán, 2007; Conati, 2010; Conati, Gertner & Vanlehn, 2002; ZapataRivera & Greer, 2003).
2.4. CONNECTING KNOWLEDGE WITH EDUCATIONAL MATERIAL
The complexity of an AEH system depends to a large extent on the
complexity of the knowledge indexing approach it uses. In the AEH
literature, indexing denotes the process of connecting domain knowledge
with educational content, i.e, specifying a set of underlying KE for every
page or fragment of educational content. This process is very similar to
traditional indexing of a page using a set of keywords. The literature
distinguishes four aspects of indexing approaches: cardinality, granularity,
navigation, and expressive power (Brusilovsky, 2003). The first two are
most important in the context of this paper.
From the cardinality aspect, there are essentially two different cases:
single KE indexing where each fragment of educational material is related
to one and only one domain model concept, and multi-concept indexing
where each fragment can be related to many concepts. Single KE indexing
is simpler and more intuitive for the authors. Multi concept indexing is more
powerful, but it makes the system more complex and requires more skilled
authoring teams.
Expressive power concerns the amount of information that the authors
can associate with every link between a concept and a page. Of course, the
most important information is the very presence of the link. This case is
called flat indexing and it is used in the majority of existing systems. Still,
some systems with a large hyperspace and advanced adaptation techniques
want to associate more information with every link by using roles and/or
weights. Assigning a role to a link helps to distinguish several kinds of
connections between concepts and pages. For example, some systems want
to distinguish between a case where a page provides an introduction, a core
explanation or a summary of a KE and a case where it provides only a core
explanation of the KE (Brusilovsky, 2000) or even some domain-specific
aspects of a KE (Brusilovsky & Cooper, 2002). Other systems use the
prerequisite role to mark the case where the KE is not presented on a page,
but it is required to understand it (Brusilovsky et al., 1998; Holden, 2003).
Existing AH systems suggest various ways of indexing that differ in all
aspects listed above. However, all this variety can be described in terms of
three basic approaches that are explored in the remaining part of this paper.
Systems using the same indexing approach have similar hyperspace
structure and share specific adaptation techniques that are based on this
structure. Thus, the indexing approach selected by developers to a large
extent defines the functionality of an AEH system.
3. Concept-Based Hyperspace: The Case of QuizGuide
The simplest approach to organizing connections between knowledge space
and hyperspace is known as concept-based hyperspace. This is the
organization approach used in an AEH system that uses single-KE indexing.
In systems with simple concept-based hyperspace, the hyperspace is built as
an exact replica of the domain model. Each KE (concept) of the domain
model is represented by exactly one node of the hyperspace, while the links
between the KE constitute main paths between hyperspace nodes. This
approach was quite popular among early AEH systems (Brusilovsky, Pesin
& Zyryanov, 1993; Hohl et al., 1996). Its current use is limited to
developing
encyclopedically
structured
learning
material
such
as
encyclopedias (Bontcheva & Wilks, 2005; Milosavljevic, 1997) and
glossaries (Brusilovsky et al., 1998; Weibelzahl & Weber, 2003). For other
kinds of practical AEH systems, multiple pages of educational material can
be created to teach the same domain model concept.
Figure 2. An enhanced concept-based hyperspace
A typical AEH system with rich content and single-concept indexing uses
an enhanced concept-based hyperspace design approach. With this design
approach, multiple pages describing the same concept are connected to this
concept in both the information space and hyperspace. Each concept has a
corresponding “hub” page in the hyperspace. The concept hub page is
connected by links to all educational hypertext pages related to this concept.
The links can be typed and weighted (Papanikolaou et al., 2003), although it
is not necessary for using the approach. The student can navigate between
hub concept pages along conceptual links and from hub pages to the pages
with educational material. An even faster approach to navigate to specific
KE and associated educational content can be provided by a visual
representation of the domain model (also known as a domain map), which is
used in AEH systems such as AES-CS (Triantafillou, Pomportis &
Demetriadis, 2003). The enhanced concept-based hyperspace approach was
used to create relatively large AEH systems with quite straightforward
structure, and allows for a number of adaptation techniques (Kavcic, 2004;
Papanikolaou et al., 2003; Steinacker et al., 2001).
Either form of concept-based hyperspace design approach provides
excellent opportunities for adaptive navigation support technologies such as
link annotation. For example, ISIS-Tutor (Brusilovsky & Pesin, 1998),
InterBook (Brusilovsky et al., 1998), and INSPIRE (Papanikolaou et al.,
2003) used annotated links to the concept hub page featuring special font
colors and icons to express the current educational state of the concept (not
known, known, well known). ISIS-Tutor (Brusilovsky & Pesin, 1998), AESCS (Triantafillou et al., 2003), ELM-ART (Weber & Brusilovsky, 2001),
and a number of other systems use annotation to show that a concept page is
not ready to be learned (i.e., its prerequisite concepts are not yet learned).
Hiding technology can be used to hide links to pages representing KEs,
which have prerequisites not yet learned (Brusilovsky & Pesin, 1998;
Kavcic, 2004) or which do not belong to the current educational goal
(Brusilovsky et al., 1998; Papanikolaou et al., 2003).
A good example of a practical system with enhanced concept-based
hyperspace is QuizGuide (Brusilovsky, Yudelson & Sosnovsky, 2004), an
adaptive front-end to a collection of interactive self-assessment questions in
the domain of C programming. The domain model in QuizGuide was
formed by 22 topics such as variables, constants or character processing. In
contrast to more traditionally used concepts, topics are coarse-grain
knowledge elements: each topic covers a relatively large fraction of domain
knowledge. QuizGuide topics were connected by prerequisite relationships
forming a network domain model. The educational content in the system
was formed by a set of more than 40 programming quizzes (each comprised
of several questions). Each quiz was classified under one of the domain
topics. Most of the topics have several quizzes associated with them, thus
forming a clean example of enhanced concept-based hyperspace, as shown
in Fig. 2.
The topic-level domain model was made visible in the QuizGuide
interface (Fig. 3) in the form of a linear topic map. Each topic name works
as a link. When a student clicks on the link, the topic opens and expands the
links to quizzes available for this topic. A click on a quiz link loads the first
question in the quiz presentation area. A click on an opened topic collapses
the list of topic questions.
Figure 3. (a) Links to topics in QuizGuide Interface were annotated with adaptive targetarrows icons displaying educational states of the topics.
(b) goal adaptation is shown by the color of the target and knowledge adaptation is indicated
by the number of errors.
Adaptive navigation support is provided in the quiz navigation area
thorugh adaptive icons shown to the left of each topic. QuizGuide adapts to
the most critical characteristics of the user: the knowledge level and the
learning goal. To reflect both the goal and knowledge relevance of each
topic in one icon, QuizGuide uses the “target-arrow” abstraction (Fig. 3).
The number of arrows in the target reflects the level of knowledge the
student has acquired on the topic: the more arrows the target has, the higher
the level of knowledge. The intensity of the target’s color shows the
relevance of the topic to the current learning goal: the more intense the color
is, the more relevant the topic. Current topics are indicated by the bright
blue targets and their direct prerequisites are indicated by dimmer blue
targets and so on. Topics that are not ready to be studied are annotated with
the crossed target. In total, there are four levels of knowledge (from zero to
three arrows) and four levels of goal relevance (not-ready, important, lessimportant and non-important). Since the student goals and knowledge are
constantly changing, different icons will be shown practically each time the
student accesses QuizGuide. To reflect changes in the user model that
happened during the same session, the student can click on the refresh icon.
Despite a relatively simple hyperspace structure and adaptation
approach, the navigation support provided by QuizGuide resulted in a
remarkable impact on student performance and motivation to work with the
system. In comparison with QuizPACK (Brusilovsky & Sosnovsky, 2005b),
an earlier version of the system which provided access to the same quizzes
with no navigation support, the average knowledge gain (a difference
between post-test and pre-test results on a 10-point test) for the students
using QuizGuide increased from 5.1 to 6.5. By guiding students to the right
topics at the right time, the system caused a significant increase in the
percentage of correctly answered questions from 35.6% to 44.3%
(Brusilovsky & Sosnovsky, 2005a). Most remarkable, however, was an
increase in the students’ interest in working with the system. The number of
attempts, the percentage of students using the system actively, and the
percentage of attempted topics increased significantly (Brusilovsky &
Sosnovsky, 2005a). The remarkable effects of QuizGuide on student
performance and motivation were discovered first in 1994 and confirmed in
several other studies (Brusilovsky, Sosnovsky & Yudelson, 2009).
Moreover, a re-implementation of QuizGuide’s adaptive navigation support
approach for SQL (Sosnovsky et al., 2008) and Java programming (Hsiao,
Sosnovsky & Brusilovsky, 2009) confirmed this impact in two other
domains.
4. Page Indexing: The Case of InterBook
The page indexing approach is typically used in cases when the volume of
educational content is relatively large and when it is desirable to increase the
precision of user modeling using finer-grained KE (which are most
frequently referred to as concepts). In these cases, page indexing (the most
straightforward implementation of multi-concept indexing) becomes very
attractive. With this approach, the whole hypermedia page (node) is indexed
with domain model concepts. In other words, links are created between a
page and each concept that is related to the content of the page (as shown in
Fig. 1). The simplest indexing approach is flat content-based indexing,
where a concept is included in a page index if some part of this page
presents the piece of knowledge corresponding to the concept (Brusilovsky
& Pesin, 1998; Henze & Nejdl, 2001). A more general – but less often used
– way to index the pages is to add the role for each concept in the page
index (role-based indexing). The most popular role is “prerequisite”: a
concept is included in a page index if a student has to know this concept to
understand the content of the page (Brusilovsky et al., 1998; De Bra, 1996;
Holden, 2003). Other roles can be used to specify the kind of contribution
that the page is providing to learning this concept (introduction, main
presentation, example, etc). Weights also can be used in multi-concept page
indexing to show how much the page contributes to learning the concept
(De Bra et al., 2002b).
A good example of the page indexing approach is provided by InterBook
(Brusilovsky et al., 1998), one of the first authoring systems for developing
AEH. ACT-R allowed the authors to create a domain-based bookshelf
containing a set of electronic textbooks on the same subject. All books on
the same bookshelf were indexed by concepts from the domain model
associated with this bookshelf using the page indexing approach. For
example, each section (page) of each textbook was connected to all concepts
related to that section. The original version of InterBook supported rolebased indexing with two roles: a concept can be either a prerequisite or an
outcome of a page. The domain model also defined the structure for an
overlay student model. As an authoring system, InterBook allowed
flexibility in defining thresholds for the different states of domain
knowledge; however, almost all AEH systems produced with InterBook
distinguished four states of student knowledge of a concept: "unknown",
"known" (learning started), "learned" and "well-learned".
The hyperspace of each bookshelf was formed by a set of electronic
textbooks and a bookshelf glossary. Textbooks were hierarchically
structured into units of different levels: chapters, sections, and subsections.
As explained above, each of these units was indexed with prerequisite and
outcome concepts. Unless hidden by settings, this indexing was clearly
visible on the border of the textbook page of InterBook (Fig. 4).
Figure 4. A textbook page and a glossary page in InterBook. Links to textbook sections are
annotated with colored bullets indicating educational states of the pages. Links to glossary
pages (which represent one concept each) are annotated with checkmarks of different sizes
indicating the current knowledge level of the explained concept.
The glossary was simply the visualized domain network. Each node of
the domain network was represented by a glossary page with links between
domain model concepts serving as navigation paths between corresponding
glossary pages. Thus, the structure of the glossary resembled the pedagogic
structure of the domain knowledge. In addition to providing a description of
a concept, each glossary page provided links to all of the book sections
which introduced or required the concept (Fig. 4). This means that the
glossary integrated traditional features of an index and a glossary. Vice
versa, concept names mentioned in the text or on the border of textbook
pages served as links to glossary pages.
The hypertext structuring approach supported by InterBook produced a
rich interlinking space with many links both within the textbook and
glossary components and between these components. To help guide users to
the most appropriate information in this multitude of links, InterBook used
two types of link annotation. Links to glossary pages were annotated with
checkmark icons of several sizes: the more knowledge of this concept
registered in the student model, the larger the size of the annotating
checkmark. Links to book sections were annotated with bullet icons of three
different colors. The bullet color (and the link font) indicated the current
educational state of the section, which was determined through tracking of
user reading. White bullets indicated pages with already learned outcome
concepts. Green bullets indicated, “ready to be learned” pages (some new
outcome concepts, but all prerequisite concepts learned already). Red bullets
marked those pages which the system considered “not ready to be learned”
(some prerequisite concepts were not yet learned). The icon and the font of
each link presented to the student were computed dynamically from the
individual student model. The goal of the latter approach was to guide the
users to interesting “ready to be learned” pages, while discouraging them
from spending too much time on “already learned” or “not ready to be
learned” pages. To provide additional guidance, the educational state of the
current page was shown by a bar of the corresponding color at the top of the
page. Needless to say, these link and text annotations were generated
dynamically taking into account the current state of individual student
knowledge.
While the adaptive navigation support provided in InterBook was
relatively simple, it had a significant impact on student navigation and
learning (Brusilovsky & Eklund, 1998). It increased student non-sequential
navigation (i.e., use of links beyond “back” and “continue”) and helped
students who followed the system’s guidance to gain better knowledge of
the subject. The prerequisite-based “traffic light” annotation approach
introduced originally in ELM-ART (Weber & Brusilovsky, 2001) and
popularized by InterBook, was later successfully applied in a number of
other systems (Carmona et al., 2002; Henze & Nejdl, 2001; Kavcic, 2004).
5. Fragment Indexing: The Case of ADAPTS
Fragment indexing is still a relatively rare indexing approach, but it is the
most precise one. The idea of the approach is to divide the content of each
hypermedia page into a set of fragments and to index some (or even all) of
these fragments with domain model concepts, which are related to the
content of these fragments. Similar to the page indexing approach, it can be
used even with unstructured vector domain models. The difference is that
indexing is done on a more fine-grained level. Generally, multi-concept
indexing is used. With smaller fragments, it is often possible to use exactly
one concept to index a fragment. In both cases, the fragment indexing
approach gives the system more precise knowledge about the content of the
page: the system knows what is presented in each indexed fragment. This
knowledge can be effectively used for advanced adaptive presentations.
Depending on the level of user knowledge about the concepts presented in a
particular fragment, the system can hide the fragment from the user (De Bra
& Calvi, 1998; Stern & Woolf, 2000), shade it (Hothi et al., 2000), or
choose one of several alternative ways to present it (Beaumont, 1994). One
of the problems in fragment-based content adaptation, especially in its
versions which hide some part of the page from users, is the lack of control
from the user side. In case of user modeling or adaptation errors, a user may
miss some valuable information without knowing of its existence. Several
approaches were suggested to return ultimate control over the process to the
user. For example, Kay (2006) argues for scrutable content adaptation
where a user can opt to see all content along with an explanation of which
parts were hidden and why. Tsandilas and schraefel (2004) suggest sliders
as a way for the user to control fragment adaptation. Höök (1996) explored
adaptive stretchtext – a specific kind of hypertext where both the user and
the system can decide which fragments are hidden or visible.
A good example of a system with fragment indexing and adaptive
stretchtest is ADAPTS (Brusilovsky & Cooper, 2002), a system for
workplace training and performance support developed for avionics
technicians. ADAPTS is able to guide the user through the troubleshooting
process building a plan of action adapted to the users’ knowledge. At each
step of the plan, the system uses adaptive content selection and adaptive
stretchtext to bring up the most relevant information (i.e., the information
which matches user goals and knowledge) from gigabytes of information
stored in an interactive
electronic
technical
manual
(IETM).
The goal of this
information is to help the user in performing this step and to expand his
knowledge (Fig. 5).
As in other AEH systems, the key to the intelligent performance of
ADAPTS is the domain model. ADAPTS uses a standard concept network
approach to domain modeling; however, due to the complexity of the
domain, its domain network is very large. The network is formed by two
main types of domain concepts: a component and a task, which form two
separate hierarchies. One hierarchy is a tree of components: from the whole
aircraft at the top, to subsystems, to sub-subsystems, down to elementary
components called addressable units. Another hierarchy is a tree of tasks:
from big diagnostic tasks that are handled by the diagnostic engine, to
subtasks, and then to elementary steps. The two hierarchies are tightly
interconnected because each task is connected with all components involved
in performing the task.
Figure 5. When presenting supporting information for a troubleshooting step, ADAPTS uses
the strerchtext approach (right): depending on user goal and knowledge, fragments can be
shown or hidden; however, the user can override system’s selection.
To support the user in performing a diagnostic task, ADAPTS uses rich
content stored in the IETM database. In addition to textual documents and
diagrams, the rich content includes various pieces of multimedia: color
photos, training videos, animations, and simulations. Moreover, the rich
content includes variations of the same information fragments oriented to
users with different levels of experience. One of the functions of ADAPTS
is to find pieces of the rich content that are relevant to the selected subtask,
and to adaptively present it to the user. To deal with large volumes of rich
content, ADAPTS uses a very elaborate indexing approach, which is
explained in detail in (Brusilovsky & Cooper, 2002). In addition to other
types of indexing, ADAPTS uses role-based indexing with components.
Conceptually, this means that each fragment of the rich content is linked by
typed (categorized) links with all components involved in this fragment. The
type of link indicates the kind of involvement (i.e., its role). For example, a
piece of video that shows how to remove a component is indexed with a
component-role pair (component ID, role=removal). Similarly, a figure that
shows the location of a component is indexed with a component-role pair
(component ID, role=location).
To match the complexity of the domain model and content indexing,
ADAPTS uses a layered multi-aspect overlay user model. A technician’s
experience with a concept can be judged on many aspects, each weighted to
indicate its relative influence on the decision. The user model independently
accumulates several aspects (roles) of the experience as well as the
knowledge of each technician about each concept as defined in the domain
model. From this record, ADAPTS uses a weighted polynomial to estimate
the proficiency of a user in locating, operating, and repairing equipment or
performing each step of a recommended procedure. The weighting of
aspects can be adjusted for different individuals. Factors measured in the
ADAPTS prototype include whether and how often a technician has
reviewed, observed, simulated, expressed understanding (self- evaluation),
previously worked on, or received certification on specific equipment or
procedures.
6. Adaptive Educational Hypermedia in a Broader
Context
The paper provided a brief overview of adaptive educational
hypermedia. As shown by multiple examples cited in the paper, AEH
technology is rich and flexible. It supports a range of personalization
scenarios and offers multiple ways to guide a student to the most relevant
learning context – presentation, examples, problems, etc. While working
well in multiple contexts, AEH is not a silver bullet and it has to be applied
with an understanding of its limitations. To start with, AEH needs to work
with a hyperspace. Hyperspace provides the best fit for educational
applications, which already use hypertext to present various educationoriented information (i.e., educational encyclopedia) or to provide access to
rich learning content (i.e., a typical Web-based education system). It is also
a good choice for any educational system that needs to operate with a large
number of information items, examples, or tasks. Even if this information is
not yet hyperlinked, it is typically not hard to structure it as a hyperspace
and AEH technologies can provide additional help by offering semantic
links. At the same time, AEH is just one of many kinds of adaptive
educational systems (Shute & Zapata-Rivera, 2010). AEH provides neither a
step-by-step problem solving support as many ITS do, nor tools for
groupwork or collaboration as collaborative learning systems. It means that
a really versatile educational and training system should not be limited to
AEH technology alone, but should wisely use a combination of technologies
to support multiple needs of students and trainees. We hope that this book as
a whole provides a well-balanced overview of many technologies and will
enable the designers of educational and training systems to create rich and
balanced systems in which AEH serves as one of the primary components.
7. References
Beaumont, I. (1994) User modeling in the interactive anatomy tutoring system
ANATOM-TUTOR. User Modeling and User-Adapted Interaction 4 (1), 21-45.
Bontcheva, K. and Wilks, Y. (2005) Tailoring automatically generated hypertext.
User Modeling and User Modeling and User Adapted Interaction 15 (1-2), 135168.
Boyle, C. and Encarnacion, A. O. (1994) MetaDoc: an adaptive hypertext reading
system. User Modeling and User-Adapted Interaction 4 (1), 1-19.
Brajnik, G., Guida, G., and Tasso, C. (1987) User modeling in intelligent
information retrieval. Information Processing and Management 23 (4), 305-320.
Brecht, B. J., McCalla, G., and Greer, J. (1989) Planning the content of instruction.
In: D. Bierman, J. Breuker and J. Sandberg (eds.) Proceedings of 4-th
International Conference on AI and Education, Amsterdam, 24-26 May 1989,
Amsterdam, IOS, pp. 32-41.
Brusilovsky, P. (1992a) A framework for intelligent knowledge sequencing and task
sequencing. In: C. Frasson, G. Gauthier and G. McCalla (eds.) Proceedings of
Second International Conference on Intelligent Tutoring Systems, ITS'92,
Montreal, Canada, June 10-12, 1992, Springer-Verlag, pp. 499-506.
Brusilovsky, P. (1993) Student as user: Towards an adaptive interface for an
intelligent learning environment. In: P. Brna, S. Ohlsson and H. Pain (eds.)
Proceedings of AI-ED'93, World Conference on Artificial Intelligence in
Education, Edinburgh, 23-27 August 1993, AACE, pp. 386-393.
Brusilovsky, P. (1996a) Adaptive hypermedia, an attempt to analyze and generalize.
In: P. Brusilovsky, P. Kommers and N. Streitz (eds.): Multimedia, Hypermedia,
and Virtual Reality. Lecture Notes in Computer Science, Vol. 1077, Berlin:
Springer-Verlag, pp. 288-304.
Brusilovsky, P. (1996b) Methods and techniques of adaptive hypermedia. User
Modeling and User-Adapted Interaction 6 (2-3), 87-129.
Brusilovsky, P. (2000) Concept-based courseware engineering for large scale Webbased education. In: G. Davies and C. Owen (eds.) Proceedings of WebNet'2000,
World Conference of the WWW and Internet, San Antonio, TX, Oct. 30 - Nov. 4,
2000, AACE, pp. 69-74.
Brusilovsky, P. (2001) Adaptive hypermedia. User Modeling and User Adapted
Interaction 11 (1/2), 87-110.
Brusilovsky, P. (2003) Developing Adaptive Educational Hypermedia Systems:
From Design Models to Authoring Tools. In: T. Murray, S. Blessing and S.
Ainsworth (eds.): Authoring Tools for Advanced Technology Learning
Environments: Toward cost-effective adaptive, interactive, and intelligent
educational software. Kluwer: Dordrecht, pp. 377-409.
Brusilovsky, P. (2004) Adaptive Educational Hypermedia: From generation to
generation. In: Proceedings of 4th Hellenic Conference on Information and
Communication Technologies in Education, Athens, Greece, September 29 October 3, 2004, pp. 19-33.
Brusilovsky, P. (2007) Adaptive navigation support. In: P. Brusilovsky, A. Kobsa
and W. Neidl (eds.): The Adaptive Web: Methods and Strategies of Web
Personalization. Lecture Notes in Computer Science, Vol. 4321, Berlin
Heidelberg New York: Springer-Verlag, pp. 263-290.
Brusilovsky, P. and Anderson, J. (1998) ACT-R electronic bookshelf: An adaptive
system for learning cognitive psychology on the Web. In: H. Maurer and R. G.
Olson (eds.) Proceedings of WebNet'98, World Conference of the WWW,
Internet, and Intranet, Orlando, FL, November 7-12, 1998, AACE, pp. 92-97.
Brusilovsky, P., Chavan, G., and Farzan, R. (2004) Social adaptive navigation
support for open corpus electronic textbooks. In: P. De Bra and W. Nejdl (eds.)
Proceedings of Third International Conference on Adaptive Hypermedia and
Adaptive Web-Based Systems (AH'2004), Eindhoven, the Netherlands, August
23-26,
2004,
Springer-Verlag,
pp.
24-33,
also
available
at
http://www2.sis.pitt.edu/~peterb/papers/AH2004Final.pdf.
Brusilovsky, P. and Cooper, D. W. (2002) Domain, Task, and User Models for an
Adaptive Hypermedia Performance Support System. In: Y. Gil and D. B. Leake
(eds.) Proceedings of 2002 International Conference on Intelligent User
Interfaces, San Francisco, CA, January 13-16, 2002, ACM Press, pp. 23-30.
Brusilovsky, P. and Eklund, J. (1998) A study of user-model based link annotation
in educational hypermedia. Journal of Universal Computer Science 4 (4), 429448.
Brusilovsky, P., Eklund, J., and Schwarz, E. (1998) Web-based education for all: A
tool for developing adaptive courseware. In: H. Ashman and P. Thistewaite (eds.)
Proceedings of Seventh International World Wide Web Conference, Brisbane,
Australia, 14-18 April 1998, Elsevier Science B. V., pp. 291-300.
Brusilovsky, P. and Henze, N. (2007) Open corpus adaptive educational
hypermedia. In: P. Brusilovsky, A. Kobsa and W. Neidl (eds.): The Adaptive
Web: Methods and Strategies of Web Personalization. Lecture Notes in Computer
Science, Vol. 4321, Berlin Heidelberg New York: Springer-Verlag, pp. 671-696.
Brusilovsky, P. and Millán, E. (2007) User models for adaptive hypermedia and
adaptive educational systems. In: P. Brusilovsky, A. Kobsa and W. Neidl (eds.):
The Adaptive Web: Methods and Strategies of Web Personalization. Lecture
Notes in Computer Science, Vol. 4321, Berlin Heidelberg New York: SpringerVerlag, pp. 3-53.
Brusilovsky, P. and Pesin, L. (1998) Adaptive navigation support in educational
hypermedia: An evaluation of the ISIS-Tutor. Journal of Computing and
Information Technology 6 (1), 27-38.
Brusilovsky, P., Pesin, L., and Zyryanov, M. (1993) Towards an adaptive
hypermedia component for an intelligent learning environment. In: L. J. Bass, J.
Gornostaev and C. Unger (eds.) Proceedings of 3rd International Conference on
Human-Computer Interaction, EWHCI'93, Berlin, August 3-7, 1993, SpringerVerlag, pp. 348-358.
Brusilovsky, P., Schwarz, E., and Weber, G. (1996a) A tool for developing adaptive
electronic textbooks on WWW. In: H. Maurer (ed.) Proceedings of WebNet'96,
World Conference of the Web Society, San Francisco, CA, October 15-19, 1996,
AACE, pp. 64-69.
Brusilovsky, P., Schwarz, E., and Weber, G. (1996b) ELM-ART: An intelligent
tutoring system on World Wide Web. In: C. Frasson, G. Gauthier and A. Lesgold
(eds.) Proceedings of Third International Conference on Intelligent Tutoring
Systems, ITS-96, Montreal, Canada, June 12-14, 1996, Springer Verlag, pp. 261269, also available at http://www.contrib.andrew.cmu.edu/~plb/ITS96.html.
Brusilovsky, P. and Sosnovsky, S. (2005a) Engaging students to work with selfassessment questions: A study of two approaches. In: Proceedings of 10th
Annual Conference on Innovation and Technology in Computer Science
Education, ITiCSE'2005, Monte de Caparica, Portugal, June 27-29, 2005, ACM
Press,
pp.
251-255,
also
available
at
http://www2.sis.pitt.edu/~peterb/papers/ITICSE05.pdf.
Brusilovsky, P. and Sosnovsky, S. (2005b) Individualized Exercises for SelfAssessment of Programming Knowledge: An Evaluation of QuizPACK. ACM
Journal on Educational Resources in Computing 5 (3), Article No. 6.
Brusilovsky, P., Sosnovsky, S., and Yudelson, M. (2005) Ontology-based
framework for user model interoperability in distributed learning environments.
In: G. Richards (ed.) Proceedings of World Conference on E-Learning, E-Learn
2005, Vancouver, Canada, October 24-28, 2005, AACE, pp. 2851-2855.
Brusilovsky, P., Sosnovsky, S., and Yudelson, M. (2009) Addictive links: The
motivational value of adaptive link annotation. New Review of Hypermedia and
Multimedia 15 (1), 97-118.
Brusilovsky, P., Yudelson, M., and Sosnovsky, S. (2004) An adaptive E-learning
service for accessing Interactive examples. In: J. Nall and R. Robson (eds.)
Proceedings of World Conference on E-Learning, E-Learn 2004, Washington,
DC, November 1-5, 2004, AACE, pp. 2556-2561.
Brusilovsky, P. L. (1992b) Student models and flexible programming course
sequencing. In: Proceedings of ICCAL'92, 4-th International Conference on
Computers and Learning, Wolfville, Canada, June 17-20, 1992, pp. 8-10.
Bunt, A., Carenini, G., and Conati, C. (2007) Adaptive content presentation for the
Web. In: P. Brusilovsky, A. Kobsa and W. Neidl (eds.): The Adaptive Web:
Methods and Strategies of Web Personalization. Lecture Notes in Computer
Science, Vol. 4321, Berlin Heidelberg New York: Springer-Verlag, pp. 409-432.
Carmona, C., Bueno, D., Guzmán, E., and Conejo, R. (2002) SIGUE: Making Web
Courses Adaptive. In: P. De Bra, P. Brusilovsky and R. Conejo (eds.)
Proceedings of Second International Conference on Adaptive Hypermedia and
Adaptive Web-Based Systems (AH'2002), Málaga, Spain, May 29-31, 2002,
Springer-Verlag, pp. 376-379.
Carro, R. M., Pulido, E., and Rodríguez, P. (1999) An Adaptive Driving Course
Based on HTML Dynamic Generation. In: P. D. Bra and J. Leggett (eds.)
Proceedings of WebNet'99, World Conference of the WWW and Internet,
Honolulu, HI, Oct. 24-30, 1999, AACE, pp. 171-176.
Conati, C. (2010) Student Modeling and Intelligent Tutoring Beyond Coached
Problem Solving. In: This book.
Conati, C., Gertner, A., and Vanlehn, K. (2002) Using Bayesian Networks to
Manage Uncertainty in Student Modeling. User Modeling and User-Adapted
Interaction 12 (4), 371-417.
Conejo, R., Guzman, E., and Millán, E. (2004) SIETTE: A Web-based tool for
adaptive teaching. International Journal of Artificial Intelligence in Education
14 (1), 29-61.
Conlan, O., Dagger, D., and Wade, V. (2002) Towards a standards-based approach
to e-Learning personalization using reusable learning objects. In: M. Driscoll and
T. C. Reeves (eds.) Proceedings of World Conference on E-Learning, E-Learn
2002, Montreal, Canada, October 15-19, 2002, AACE, pp. 210-217.
Dagger, D., Wade, V., and Conlan, O. (2004) A Framework for Developing
Adaptive Personalized eLearning. In: J. Nall and R. Robson (eds.) Proceedings of
World Conference on E-Learning, E-Learn 2004, Washington, DC, USA,
November 1-5, 2004, AACE, pp. 2579-2587.
Davidovic, A., Warren, J., and Trichina, E. (2003) Learning benefits of structural
example-based adaptive tutoring systems. IEEE Transactions on Education 46
(2), 241-251.
De Bra, P., Aerts, A., and Rousseau, B. (2002a) Concept Relationship Types for
AHA! 2.0. In: M. Driscoll and T. C. Reeves (eds.) Proceedings of World
Conference on E-Learning, E-Learn 2002, Montreal, Canada, October 15-19,
2002, AACE, pp. 1386-1389.
De Bra, P., Aerts, A., Smits, D., and Stash, N. (2002b) AHA! Version 2.0: More
Adaptation Flexibility for Authors. In: M. Driscoll and T. C. Reeves (eds.)
Proceedings of World Conference on E-Learning, E-Learn 2002, Montreal,
Canada, October 15-19, 2002, AACE, pp. 240-246.
De Bra, P. and Calvi, L. (1998) AHA! An open Adaptive Hypermedia Architecture.
The New Review of Hypermedia and Multimedia 4, 115-139.
De Bra, P. and Ruiter, J.-P. (2001) AHA! Adaptive hypermedia for all. In: W.
Fowler and J. Hasebrook (eds.) Proceedings of WebNet'2001, World Conference
of the WWW and Internet, Orlando, FL, October 23-27, 2001, AACE, pp. 262268.
De Bra, P. M. E. (1996) Teaching Hypertext and Hypermedia through the Web.
Journal of Universal Computer Science 2 (12), 797-804.
de La Passardiere, B. and Dufresne, A. (1992) Adaptive navigational tools for
educational hypermedia. In: I. Tomek (ed.) Proceedings of ICCAL'92, 4-th
International Conference on Computers and Learning, Berlin, June 17-20, 1992,
Springer-Verlag, pp. 555-567.
Denaux, R., Dimitrova, V., and Aroyo, L. (2005) Integrating Open User Modeling
and Learning Content Management for the Semantic Web. In: L. Ardissono, P.
Brna and A. Mitrovic (eds.) Proceedings of 10th International User Modeling
Conference, Edinburgh, Scotland, UK, July 24-29, 2005, Springer Verlag, pp. 918.
Dichev, C., Dicheva, D., and Aroyo, L. (2004) Using Topic Maps for Web-based
Education. Advanced Technology for Learning 1 (1), 1-7.
Dolog, P., Gavriloaie, R., Nejdl, W., and Brase, J. (2003) Integrating Adaptive
Hypermedia Techniques and Open RDF-Based Environments. In: Proceedings of
The Twelfth International World Wide Web Conference, WWW 2003, Budapest,
Hungary, 20-24 May, 2003, pp. 88-98.
Dolog, P. and Nejdl, W. (2007) Semantic Web Technologies for the Adaptive Web.
In: P. Brusilovsky, A. Kobsa and W. Neidl (eds.): The Adaptive Web: Methods
and Strategies of Web Personalization. Lecture Notes in Computer Science, Vol.
4321, Berlin Heidelberg New York: Springer-Verlag, pp. 697-719.
Farrell, R., Thomas, J. C., Dooley, S., Rubin, W., Levy, S., O’Donnell, R., and
Fuller, E. (2003) Learner-driven assembly of Web-based courseware. In: A.
Rossett (ed.) Proceedings of World Conference on E-Learning, E-Learn 2003,
Phoenix, AZ, USA, November 7-11, 2003, AACE, pp. 1052-1059.
Gilbert, J. E. and Han, C. Y. (1999) Arthur: Adapting Instruction to Accommodate
Learning Style. In: P. D. Bra and J. Leggett (eds.) Proceedings of WebNet'99,
World Conference of the WWW and Internet, Honolulu, HI, Oct. 24-30, 1999,
AACE, pp. 433-438.
Gonschorek, M. and Herzog, C. (1995) Using hypertext for an adaptive helpsystem
in an intelligent tutoring system. In: J. Greer (ed.) Proceedings of AI-ED'95, 7th
World Conference on Artificial Intelligence in Education, Washington, DC, 1619 August 1995, AACE, pp. 274-281.
Henze, N. (2005) Personal Readers: Personalized Learning Object Readers for the
Semantic Web. In: C.-K. Looi, G. McCalla, B. Bredeweg and J. Breuker (eds.)
Proceedings of 12th International Conference on Artificial Intelligence in
Education, AIED'2005, Amsterdam, July 18-22, 2005, IOS Press, pp. 274-281,
also
available
at
http://www.kbs.unihannover.de/Arbeiten/Publikationen/2005/aied05.pdf.
Henze, N., Naceur, K., Nejdl, W., and Wolpers, M. (1999) Adaptive hyperbooks for
constructivist teaching. Künstliche Intelligenz (4), 26-31.
Henze, N. and Nejdl, W. (1999) Student modeling for KBS Hyperbook system
using Bayesian networks, Technical report, Report, University of Hannover.
Henze, N. and Nejdl, W. (2001) Adaptation in open corpus hypermedia.
International Journal of Artificial Intelligence in Education 12 (4), 325-350.
Hockemeyer, C., Held, T., and Albert, D. (1998) RATH - A relational adaptive
tutoring hypertext WWW-environment based on knowledge space theory. In: C.
Alvegård (ed.) Proceedings of CALISCE'98, 4th International conference on
Computer Aided Learning and Instruction in Science and Engineering, Göteborg,
Sweden, June 15-17, 1998, pp. 417-423.
Hohl, H., Böcker, H.-D., and Gunzenhäuser, R. (1996) Hypadapter: An adaptive
hypertext system for exploratory learning and programming. User Modeling and
User-Adapted Interaction 6 (2-3), 131-156.
Holden, S. (2003). Architecture for scrutable adaptive hypermedia teaching from
diverse document collection (Ph.D. Thesis). The University of Sydney.
Hoog, R. d., Wielinga, B., Kabel, S., Anjewierden, A., Verster, F., Barnard, Y.,
PaoloDeLuca, Desmoulins, C., and Riemersma, J. (2002) Re-using technical
manuals for instruction: document analysis in the IMAT project. In: Y. Barnard
(ed.) Proceedings of Workshop on integrating technical and training
documentation held in conjuction with ITS'02 conference, San Sebastian, Spain,
June 3, 2002, pp. 15-25.
Höök, K., Karlgren, J., Wærn, A., Dahlbäck, N., Jansson, C. G., Karlgren, K., and
Lemaire, B. (1996) A glass box approach to adaptive hypermedia. User Modeling
and User-Adapted Interaction 6 (2-3), 157-184.
Hothi, J., Hall, W., and Sly, T. (2000) A study comparing the use of shaded text and
adaptive navigation support in adaptive hypermedia. In: P. Brusilovsky, O. Stock
and C. Strapparava (eds.) Proceedings of Adaptive Hypermedia and Adaptive
Web-based systems, Berlin, August 28-30, 2000, Springer-Verlag, pp. 335-342.
Hsiao, I.-H., Sosnovsky, S., and Brusilovsky, P. (2009) Adaptive Navigation
Support for Parameterized Questions in Object-Oriented Programming. In: U.
Cress, V. Dimitrova and M. Specht (eds.) Proceedings of 4th European
Conference on Technology Enhanced Learning (ECTEL 2009), Nice, France,
September 29- October 2, 2009, Springer-Verlag, pp. 88-98.
Hsiao, I.-H., Sosnovsky, S., and Brusilovsky, P. (2010) Guiding Students to the
Right Questions: Adaptive Navigation Support in an E-learning System for Java
Programming. Journal of Computer Assisted Learning, in press.
Jacquiot, C., Bourda, Y., and Popineau, F. (2004) GEAHS: A Generic Educational
Adaptive Hypermedia System. In: L. Cantoni and C. McLoughlin (eds.)
Proceedings of ED-MEDIA'2004 - World Conference on Educational
Multimedia, Hypermedia and Telecommunications, Lugano, Switzerland, June
21-26, 2004, AACE, pp. 571-578.
Kaplan, C., Fenwick, J., and Chen, J. (1993) Adaptive hypertext navigation based
on user goals and context. User Modeling and User-Adapted Interaction 3 (3),
193-220.
Kavcic, A. (2004) Fuzzy User Modeling for Adaptation in Educational
Hypermedia. IEEE Transactions on Systems, Man, and Cybernetics 34 (4), 439449.
Kay, J. (2006) Scrutable adaptation: Because we can and must. In: V. Wade, H.
Ashman and B. Smyth (eds.) Proceedings of 4th International Conference on
Adaptive Hypermedia and Adaptive Web-Based Systems (AH'2006), Dublin,
Ireland, June 21-23, 2006, Springer Verlag, pp. 11-19.
Kay, J. and Kummerfeld, B. (2010) Lifelong learner modeling In: This book.
Knutov, E., De Bra, P., and Pechenizkiy, M. (2009) AH 12 years later: a
comprehensive survey of adaptive hypermedia methods and techniques. New
Review of Hypermedia and Multimedia 15 (1), 5-38.
Laroussi, M. and Benahmed, M. (1998) Providing an adaptive learning through the
Web case of CAMELEON: Computer Aided MEdium for LEarning on Networks.
In: C. Alvegård (ed.) Proceedings of CALISCE'98, 4th International conference
on Computer Aided Learning and Instruction in Science and Engineering,
Göteborg, Sweden, June 15-17, 1998, pp. 411-416.
Milosavljevic, M. (1997) Augmenting the user's knowledge via comparison. In: A.
Jameson, C. Paris and C. Tasso (eds.) Proceedings of 6th International
Conference on User Modeling, UM97, Chia Laguna, Sardinia, Italy, June 2-5,
1997, SpringerWienNewYork, pp. 119-130.
Mitrovic, A. and Devedzic, V. (2004) A Model of Multitutor Ontology-based
Learning Environments. Continuing Engineering Education and Life-Long
Learning 14 (3), 229-245.
Mödritscher, F., García Barrios, V. M., and Gütl, C. (2004) Enhancement of
SCORM to support adaptive E-Learning within the Scope of the Research Project
AdeLE. In: J. Nall and R. Robson (eds.) Proceedings of World Conference on ELearning, E-Learn 2004, Washington, DC, USA, November 1-5, 2004, AACE,
pp. 2499-2505.
Morimoto, Y., Ueno, M., Kikukawa, I., Yokoyama, S., and Miyadera, Y. (2007)
SALMS: SCORM-compliant Adaptive LMS. In: T. Bastiaens and S. Carliner
(eds.) Proceedings of World Conference on E-Learning, E-Learn 2007, Quebec
City, Canada, October 15-19, 2007, AACE, pp. 7287-7296.
Murray, T., Piemonte, J., Khan, S., Shen, T., and Condit, C. (2000) Evaluating the
need for intelligence in an adaptive hypermedia system. In: G. Gauthier, C.
Frasson and K. VanLehn (eds.) Proceedings of 5th International Conference on
Intelligent Tutoring Systems (ITS'2000), Berlin, June 21-23, 2000, SpringerVerlag, pp. 373-382.
Oberlander, J., O'Donell, M., Mellish, C., and Knott, A. (1998) Conversation in the
museum: experiments in dynamic hypermedia with the intelligent labeling
explorer. The New Review of Multimedia and Hypermedia 4, 11-32.
Oda, T., Satoh, H., and Watanabe, S. (1998) Searching deadlocked Web learners by
measuring similarity of learning activities. In: Proceedings of Workshop
"WWW-Based Tutoring" at 4th International Conference on Intelligent Tutoring
Systems (ITS'98), San Antonio, TX, August 16-19, 1998, also available at
http://www.sw.cas.uec.ac.jp/~watanabe/conference/its98workshop1.ps.
Papanikolaou, K. A., Grigoriadou, M., Kornilakis, H., and Magoulas, G. D. (2003)
Personalising the interaction in a Web-based Educational Hypermedia System:
the case of INSPIRE. User Modeling and User Adapted Interaction 13 (3), 213267.
Paris, C. L. (1988) Tailoring object description to a user's level of expertise.
Computational Linguistics 14 (3), 64-78.
Pérez, T., Gutiérrez, J., and Lopistéguy, P. (1995) An adaptive hypermedia system.
In: J. Greer (ed.) Proceedings of AI-ED'95, 7th World Conference on Artificial
Intelligence in Education, Washington, DC, 16-19 August 1995, AACE, pp. 351358.
Rey-López, M., Brusilovsky, P., Meccawy, M., Díaz-Redondo, R. P., FernándezVilas, A., and Ashman, H. (2008) Resolving the Problem of Intelligent Learning
Content in Learning Management Systems. International Journal on E-Learning
7 (3), 363-381.
Sanrach, C. and Grandbastien, M. (2000) ECSAIWeb: A Web-based authoring
system to create adaptive learning systems. In: P. Brusilovsky, O. Stock and C.
Strapparava (eds.) Proceedings of Adaptive Hypermedia and Adaptive Webbased Systems, AH2000, Trento, Italy, August 28-30, 2000, Springer-Verlag, pp.
214-226.
Schneider-Hufschmidt, M., Kühme, T., and Malinowski, U. (eds.) (1993) Adaptive
user interfaces: Principles and practice. Human Factors in Information
Technology, Amsterdam: North-Holland.
Schöch, V., Specht, M., and Weber, G. (1998) "ADI" - an empirical evaluation of a
tutorial agent. In: T. Ottmann and I. Tomek (eds.) Proceedings of EDMEDIA/ED-TELECOM'98 - 10th World Conference on Educational Multimedia
and Hypermedia and World Conference on Educational Telecommunications,
Freiburg, Germany, June, 20-25, 1998, AACE, pp. 1242-1247.
Shute, V. J. and Zapata-Rivera, D. (2010) Adaptive Educational Systems. In: This
book.
Soller, A. (2007) Adaptive support for distributed collaboration. In: P. Brusilovsky,
A. Kobsa and W. Neidl (eds.): The Adaptive Web: Methods and Strategies of Web
Personalization. Lecture Notes in Computer Science, Vol. 4321, Berlin
Heidelberg New York: Springer-Verlag, pp. 573-595.
Sosnovsky, S., Brusilovsky, P., Lee, D. H., Zadorozhny, V., and Zhou, X. (2008)
Re-assessing the Value of Adaptive Navigation Support in E-Learning. In: W.
Nejdl, J. Kay, P. Pu and E. Herder (eds.) Proceedings of 5th International
Conference on Adaptive Hypermedia and Adaptive Web-Based Systems
(AH'2008), Hannover, Germany, July 29-August 1, 2008, Springer Verlag, pp.
193-203, also available at http://dx.doi.org/10.1007/978-3-540-70987-9_22.
Specht, M. and Klemke, R. (2001) ALE - Adaptive Learning Environment. In: W.
Fowler and J. Hasebrook (eds.) Proceedings of WebNet'2001, World Conference
of the WWW and Internet, Orlando, FL, October 23-27, 2001, AACE, pp. 11551160.
Specht, M., Kravcik, M., Klemke, R., Pesin, L., and Hüttenhain, R. (2002) Adaptive
Learning Environment (ALE) for Teaching and Learning in WINDS. In:
Proceedings of Second International Conference on Adaptive Hypermedia and
Adaptive Web-Based Systems (AH'2002), Berlin, May 29-31, 2002, SpringerVerlag, pp. 572-581.
Specht, M. and Oppermann, R. (1998) ACE - Adaptive Courseware Environment.
The New Review of Hypermedia and Multimedia 4, 141-161.
Steinacker, A., Faatz, A., Seeberg, C., Rimac, I., Hörmann, S., Saddik, A. E., and
Steinmetz, R. (2001) MediBook: Combining semantic networks with metadata
for learning resources to build a Web based learning system. In: Proceedings of
ED-MEDIA'2001 - World Conference on Educational Multimedia, Hypermedia
and Telecommunications, Tampere, Finland, June 25-30, 2001, AACE, pp. 17901795.
Steinacker, A., Seeberg, C., Rechenberger, K., Fischer, S., and Steinmetz, R. (1999)
Dynamically generated tables of contents as guided tours in adaptive hypermedia
systems. In: P. Kommers and G. Richards (eds.) Proceedings of ED-MEDIA/EDTELECOM'99 - 11th World Conference on Educational Multimedia and
Hypermedia and World Conference on Educational Telecommunications, Seattle,
WA, AACE, pp. 640-645.
Stern, M. K. and Woolf, B. P. (2000) Adaptive content in an online lecture system.
In: P. Brusilovsky, O. Stock and C. Strapparava (eds.) Proceedings of Adaptive
Hypermedia and Adaptive Web-based systens, Berlin, August 28-30, 2000,
Springer-Verlag, pp. 225-238.
Stock, O., Zancanaro, M., Busetta, P., Callaway, C., Krüger, A., Kruppa, M.,
Kuflik, T., Not, E., and Rocchi, C. (2007) Adaptive, intelligent presentation of
information for the museum visitor in PEACH. User Modeling and User-Adapted
Interaction 17 (3), 257-304.
Trausan-Matu, S., Maraschi, D., and Cerri, S. A. (2002) Ontology-centered
personalized presentation for knowledge extracted from the Web. In: S. A. Cerri,
G. Gouardères and F. Paraguaçu (eds.) Proceedings of 6th International
Conference on Intelligent Tutoring Systems (ITS'2002), Berlin, June 2-7, 2002,
Springer-Verlag, pp. 259-269.
Trella, M., Conejo, R., and Bueno, D. (2002) An autonomous component
architecture to develop WWW-ITS. In: P. Brusilovsky, N. Henze and E. Millán
(eds.) Proceedings of Workshop on Adaptive Systems for Web-Based Education
at the 2nd International Conference on Adaptive Hypermedia and Adaptive WebBased Systems (AH'2002), Málaga, Spain, May 28, 2002, pp. 69-80.
Triantafillou, E., Pomportis, A., and Demetriadis, S. (2003) The design and the
formative evaluation of an adaptive educational system based on cognitive styles.
Computers and Education, 87-103.
Tsandilas, T. and schraefel, M. C. (2004) Usable adaptive hypermedia systems.
New Review in Hypermedia and Multimedia 10 (1), 5.
Ueno, M. (2005) Intelligent LMS with an agent that learns from log data. In: G.
Richards (ed.) Proceedings of World Conference on E-Learning, E-Learn 2005,
Vancouver, Canada, October 24-28, 2005, AACE, pp. 2068-2074.
Vassileva, J. (1998) DCG + GTE: Dynamic Courseware Generation with Teaching
Expertise. Instructional Science 26 (3/4), 317-332.
Weber, G. and Brusilovsky, P. (2001) ELM-ART: An adaptive versatile system for
Web-based instruction. International Journal of Artificial Intelligence in
Education 12 (4), 351-384.
Weber, G., Kuhl, H.-C., and Weibelzahl, S. (2001) Developing adaptive internet
based courses with the authoring system NetCoach. In: P. D. Bra, P. Brusilovsky
and A. Kobsa (eds.) Proceedings of Third workshop on Adaptive Hypertext and
Hypermedia, Sonthofen, Germany, July 14, 2001, pp. 35-48, also available at
http://wwwis.win.tue.nl/ah2001/papers/GWeber-UM01.pdf, also available at
http://wwwis.win.tue.nl/ah2001/papers/GWeber-UM01.pdf.
Weibelzahl, S. and Weber, G. (2003) Evaluating the inference mechanism of
adaptive learning systems. In: P. Brusilovsky, A. Corbett and F. d. Rosis (eds.)
Proceedings of 9th International User Modeling Conference, Johnstown, PA, June
22-26, 2003, Springer Verlag, pp. 154-162.
Zapata-Rivera, J.-D. and Greer, J. E. (2003) Student model accuracy using
inspectable Bayesian student models. In: U. Hoppe, F. Verdejo and J. Kay (eds.)
Proceedings of AI-Ed'2003, Amsterdam, July 22-24, 2003, IOS Press, pp. 65-72.