Declarative Programming Project – Gerrymandering (ASP + Python)

This project demonstrates the use of Answer Set Programming (ASP) to model and analyze gerrymandering strategies in grid-based electoral systems.

The objective is to partition a voting region (represented as a 2D grid) into a fixed number of contiguous districts to maximize the number of seats won by a given political party.

---

📂 Repository Structure

The repository contains the following main components:

• benchmarks/ Contains 100 benchmark instances in ASP format (.lp).

• imgs/ Graphical results, plots, and heatmaps used in the report.

• report/ PDF reports containing theoretical background and experimental analysis (English and Italian versions).

• test/ Utility scripts for vote counting (count_voters.py) and debugging (debug_districting.lp).

• extra/ Notebooks and study notes from the Declarative Programming course.

• political_districting.lp The main ASP encoding for the districting problem.

• generate_benchmarks.py Python script used to generate randomized benchmark instances.

• run_experiments_python.py Python script orchestrating the solver using Clingo.

• results_python.csv Output file collecting all experiment results.

---

📄 Reports & Documentation

Theoretical background, model structure, and experimental analysis are detailed in the project reports:

• 🇬🇧 report/GerrymanderingENG.pdf – English version
• 🇮🇹 report/GerrymanderingITV.pdf – Italian version

---

👥 Contacts

For inquiries or feedback regarding this project:

• Aurora Felisari - aurora.felisari@studenti.unipr.it
• Claudio Bendini - claudio.bendini@studenti.unipr.it

---

🎓 Project Context

This work was developed as part of the Declarative Programming course at the University of Parma to explore logic-based modeling and optimization using ASP and Python.