Implementation and Visualization of Conceptual Graphs in CharGer

Implementation and Visualization of Conceptual Graphs in CharGer

Harry S. Delugach (The University of Alabama in Huntsville, Huntsville, AL, USA)
DOI: 10.4018/IJCSSA.2014070101
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Practitioners using conceptual graphs are naturally interested in tools that support its features. The CharGer editing software has been available for many years, giving the community a chance to evaluate and contribute to CharGer's ever-expanding capabilities. As result of that experience, the author discusses a few lessons learned about visualization and implementation of conceptual graphs. Features such as actors, glossary integration, standards-based interoperability and automatic layout offer opportunities to study and improve the usefulness of conceptual graphs tools.
Article Preview

Introduction

CharGer is a conceptual graph (CG) visual editor written in Java. It is intended for researchers and practitioners who want to create and edit graphs for CG knowledge bases. CharGer supports most CG features, including actors, context and (importantly) knowledge interchange. It is arguably one of the most significant tools available in the development of CGs. Its development has been informed and guided by many interested users, researchers and developers over the past 20 years.

There are other CG packages in wide use, but we will mention two others specifically, since they have user communities similar in size and scope to CharGer. They are CoGui/Cogitant (Chein & Mugnier, 2008) and Amine (Kabbaj, 2006). While there is overlap between their features and CharGer, these platforms focus more on reasoning and ontology building whereas CharGer is primarily aimed at the user interactive potential of CGs. CharGer has addressed several important issues in the visualization of CGs, while its implementation serves as a proof-of-concept platform for several features that can enhance the usefulness of CGs in the larger landscape of other knowledge representation tools and systems. In this paper we aim to highlight several issues and considerations that bear some relevance to knowledge visualization in general, as well as visualization and implementation of CGs in particular. We believe our experience with CG implementations provides some valuable lessons learned, some useful implementation strategies, and ultimately identifies some open issues in implementing CGs that will spur further discussion and exploration.

This paper assumes the reader is already reasonably familiarity with CGs. For general background on CG theory, other excellent resources are available (Chein & Mugnier, 2008; Corbett, 2003; Sowa, 2008; Sowa, 1984). For a tutorial and user’s guide for CharGer, see the CharGer User’s Guide (Delugach, 2014). The software itself is freely available at (Delugach, 2009b). 1

Brief History of CharGer

CharGer has a long history of development and use, much of it occurring in conjunction with the theoretical and formal development of conceptual graphs. It began (as do many such projects) as an experiment in understanding how people might use conceptual graphs in an interactive environment. The first parts were written in early 1997, and by 1998 the name “CharGer” was first used. (The name was chosen because the University of Alabama in Huntsville has the “Chargers” as its mascot, with both “C” and “G” in the name.) As more researchers in CGs began using it, new features were invariably requested. By 1999, CharGer version 1.7 was released.

CharGer has undergone continuous enhancement and revision since 1999. In 2001, a robust actor activation subsystem was in place. An interface to a repertory grid system was added in 2003. Wordnet integration followed in 2004. Cuts were also introduced in 2004. Compatibility with the CGIF interchange format was first attempted in 2007 and was more-or-less complete by 2013. Automatic layout was also added in 2014. A more modern editing screen was added in 2014.

As of version 4.1 in mid 2015, CharGer consists of about 50,000 lines of Java source code in 24 Java packages comprising 221 classes. Much of its current user interface was developed using GUI tools in the Netbeans IDE under Mac OS. Since CharGer was placed at the software sharing site Sourceforge.net in 2005; since then, there have been almost 10,000 downloads from over 70 countries.

Features

This section lists some of the main features of CharGer, as an illustration of the range of capabilities that a practical environment should provide in order to support knowledge visualization and analysis in CGs. A complete list of features can be found in the CharGer User’s Guide (Delugach, 2014).

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 5: 2 Issues (2017)
Volume 4: 2 Issues (2016)
Volume 3: 2 Issues (2015)
Volume 2: 2 Issues (2014)
Volume 1: 2 Issues (2013)
View Complete Journal Contents Listing