Tool Comparison

[yilinxia]

We need to review the existing work in the field and ensure we clearly highlight how our work differs.

• draco https://github.com/cmudig/draco2
• clinguin https://github.com/potassco/clinguin
• clingraph https://github.com/potassco/clingraph

[yilinxia]

• erie https://github.com/see-mike-out/erie-web
• dracoGPT https://github.com/wwwhhhccc/DracoGPT

[yilinxia]

Nodes Comparing Clingraph/Clingui with DecViz

• Flexibility Clingraph/clinguin treat graph attributes as predicates. In contrast, DecViz uses them more like variables through named arguments. This approach provides greater flexibility - users can simply specify what they want, as in graphviz, without having to fill every variable as in Clingo.
• Extensibility It also enhances extensibility - attributes such as edge color or border style can be easily added (not to mention like the general settings for graph direction, rank, rendering engine...)
• Scalability This is primarily due to the fundamental difference between the underlying engines of Logica and Clingo: pure Logica can efficiently compute large sets of facts, whereas Clingo is not designed for that purpose.
• Model Exploration Logica has introduced an embedded function for analyzing Clingo’s output, enabling model selection and exploration for further visualization within a declarative paradigm.
• Not Purely Declarative: Although Clingraph uses ASP to generate visualizations (including animations), certain settings are managed through the command line rather than through the declarative programming process e.g., --select -model=0 , --engine,

But there are some advantages for Clingraph

• User Case DecViz is specifically designed for Argumentation Frameworks (or graph visualization in general), whereas Clingraph is more focused on ASP-based examples and offers a broader range of interactive use cases.
• Animation Right now DecViz doesnt support any animations but it is cool feature to consider (i.e., visualize the reasoning process)