ClustArg

Introduction

ClustArg is a versatile tool designed with a focus on abstract Argumentation Frameworks (AFs), providing a powerful platform for analyzing and computing different argumentation semantics. The tool is intended to help researchers and practitioners explore and evaluate abstract AFs, which represent the structure of argumentation without specifying the details of individual arguments. By implementing three core semantics—conflict-free, admissible, and stable—ClustArg enables the assessment of the acceptability and coherence of arguments within these abstract frameworks. Additionally, the tool supports refinements and two distinct extension calculation processes: Breadth-First Search (BFS) and Depth-First Search (DFS), offering flexible approaches for analyzing AFs from different perspectives.

With ClustArg, users can:

• Compute the semantics of concrete AFs, providing insights into the acceptability of individual arguments and sets of arguments.
• Compute the semantics of abstract AFs, enabling the analysis of abstract argumentation structures without needing a fully instantiated set of arguments.
• Check if an abstract AF is faithful, ensuring that the abstract representation correctly reflects the structure of the argumentation.
• Compute semantic extensions and identify situations leading to spuriousness, where extensions may not align with expected logical behavior.
• Transform an AF that was initially spurious into one that is faithful, ensuring the argumentation structure is logically consistent and valid.

ClustArg is intended for researchers, developers, and anyone interested in exploring the dynamic landscape of argumentation frameworks and their semantics.

Installation

ClustArg is coded in Python and includes a requirements.txt file for easy installation of dependencies. To set up the tool, simply follow these steps:

1. Clone this repository:

git clone https://github.com/p4s3r0/argumentation-framework-clustering.git

2. Navigate into the project directory:

cd argumentation-framework-clustering

3. (Optional but recommended) Set up a virtual environment to manage dependencies:

python3 -m venv venv
source venv/bin/activate  # On Windows use: venv\Scripts\activate

4. Install the required dependencies:

pip3 install -r requirements.txt

Usage

Generate Semantics Extensions

To generate semantic extensions, the SETS program can be used. By replacing <sem> with the appropriate abbreviation (CF for conflict-free, AD for admissible, or ST for stable) and <inp> with the input path of the AF, the program can produce the desired semantics.

python3 main.py SETS <inp> -s <sem>

Determine Faithfulness/Spuriousness

Use the CHECK program to determine whether the abstract AF is faithful relative to the concrete AF. Replace <abs> with the file name of the abstract AF, <con> with the file name of the concrete AF and with the appropriate semantics. This tool enables verification of faithfulness between the abstract and concrete argumentation frameworks.

python3 main.py CHECK <abs> -c <con> -s <sem>

Create Faithful AF

Use the FAITHFUL program to transform a spurious abstract AF into a faithful AF based on the concrete AF . Additionally, replace with the corresponding semantics (i.e., CF, AD, or ST) to guide the transformation process. This ensures the resulting abstract AF aligns faithfully with the concrete AF under the specified semantics.

python3 main.py FAITHFUL <abs> -c <con> -s <sem>

Concretize specific Arguments and Generate Faithful AF

Use the CONCRETIZE program to concretize a list of arguments and include additional arguments to ensure faithfulness. Replace <abs> with the abstract AF file, <con> with the concrete AF file, <sem> with the semantic abbreviation (i.e., CF, AD, or ST), and <args> with a space-separated list of arguments. This process helps maintain consistency and faithfulness between the abstract and concrete argumentation frameworks.

python3 main.py CONCRETIZE <abs> -c <con> -s <sem> -p <args>

Optional Flags

Flag	Description	Options
f	Program to be executed	SETS, CHECK, FAITHFUL, CONCRETIZE
i	Path to the input file	-
-c	Defines the path of the file containing the concrete AF	-
-s	Defines the semantics	CF, AD, ST
-a	Specifies the algorithm	BFS, DFS
-p	Lists the arguments to be concretized separated by spaces	-
-vis	Boolean value to visualize AFs	None
-verbose	Boolean value to print additional computation steps	None
-noref	Boolean value to disable refinements	None
-exp

TODO

References and Other Works

Checking the acceptability of a set of arguments

Existential Abstraction on Argumentation Frameworks via Clustering