Visualizing Reuse: More than Meets the Eye


Joris Klerkx
(Katholieke Universiteit Leuven, Belgium
joris.klerkx@cs.kuleuven.be)

Katrien Verbert
(Katholieke Universiteit Leuven, Belgium
katrien.verbert@cs.kuleuven.be)

Erik Duval
(Katholieke Universiteit Leuven, Belgium
erik.duval@cs.kuleuven.be)



Abstract: In this paper we discuss an interactive visualization application that aims to visualize
a large repository of small reusable content components that were created by disaggregating
legacy content. The purpose of this decomposition is to produce content that can be
automatically reused in on-the-fly assemblies of new learning objects. The purpose of the
visualization application is to offer insight in the structure of the contents of the repository and
to enable access to them in an effective and efficient way.
Keywords: learning objects, reusability, learning object repositories, metadata, information
visualization
Categories: H.3.1, H.3.3, H.3.7
1 Introduction
Learning objects (LOs) are often very coarse-grained and difficult to reuse due to the
fact that they are stored in a final presentation form. This static representation is not
suitable for flexible content reuse, as the components cannot be easily accessed. In
many cases, paragraphs, images or diagrams are assembled manually by copy and
paste actions. In order to support this process in a more methodical way, we
decompose composite learning objects, and make those components available for
more flexible content reuse. This process results in a Learning Object Repository
(LOR) with a large amount of LO components and requires advanced support for
searching and finding relevant components.
In [Klerkx, et al., 04] we covered our initial research on novel access paradigms
for LORs, through Information Visualization techniques. We described the use of
tree-maps, hyperbolic trees and Venn diagrams to visualize LOR contents. In [Klerkx,
et al., 05], we presented our open and extensible information visualization framework.
This framework was developed to support our research by enabling us to rapidly
experiment with:
o case-studies that rely on a variety of data sources, with a variety of metadata
schemes like IEEE LTSC Learning Object Metadata (LOM), Dublin Core,
etc.
Proceedings of I-KNOW ’06
Graz, Austria, September 6 - 8, 2006

o visualization techniques, like tree-maps, radial trees, fisheye views,
zoomable user interfaces, etc. [Card, et al., 99].
o information visualization toolkits like Piccolo [Bederson, et al., 04], INRIA’s
Infovis [Fekete, 04], the InfoVis Cyberinfrastructure [Baumgartner, et al.,
04], etc. as we don’t want to reinvent the wheel and prefer to combine the
strengths of several toolkits.
In this paper we discuss an interactive visualization application that was created by
using this framework. The specific application discussed in this paper visualizes a
repository of small reusable content components. These components were created by
disaggregating legacy content. The purpose of the decomposition is to produce
content that can be automatically reused in on-the-fly assemblies of new learning
objects – see section 2.
This paper is structured as follows: in the next section, we offer background on
the data source and its objectives. The visualization application itself is discussed in
section 3. In section 4, we elaborate in a discussion about early evaluation and future
work. We conclude the paper in section 5.
2 The ALOCOM framework
The ALOCOM framework supports two processes: the decomposition of learning
objects into their components (text fragments, images, definitions, diagrams, tables,
examples, audio and video sequences …) as well as the automatic assembly of these
components in real-world authoring tools.
o In the disaggregation process, a composite learning object is decomposed
into its components. In the case of a slide presentation, the presentation is
disaggregated into the individual slides and each slide is further decomposed
into its images, diagrams, tables, text fragments, definitions, references…
Metadata is added to each learning object component and the component is
stored in a repository [Verbert et al., 2005].
o The aggregation process searches for components in the learning object
repository and adds them to the learning object. Since users prefer to use
authoring environments they are familiar with, this functionality has to be
integrated in ordinary authoring tools. We have currently developed a
component search plug-in for MS PowerPoint (Figure 1b) and MS Word. A
user can specify the type of component he or she is interested in (e.g. slide,
image, text fragment, table, diagram…), as well as keywords that best
describe the component. All components that satisfy the specified search
criteria are shown and the author can easily incorporate them into the
learning object he or she is working on. For more details, we refer to
[Verbert et al., 2005].

In the remainder of this section, the decomposition process and the ALOCOM
repository are described in more detail.
490 Klerkx J., Verbert K., Duval E.: Visualizing Reuse ...