AcciMaps Analysis

A comprehensive workflow for processing PDF and PNG files, extracting text, and performing graph-based community analysis. This repository includes a Jupyter Notebook for processing, installation guides, and supporting documentation.

Files Included

1. Structure.pdf
   Contains the setup instructions and detailed environment requirements for running the Jupyter Notebook.

2. Installation Guide.pdf
   Provides detailed descriptions of the cells in the notebook, their purpose, and their integration into the overall workflow.

3. PDF2_Multiple.ipynb
   A Jupyter Notebook implementing the pipeline for processing PDF and PNG files, including text extraction, graphical analysis, and clustering.

Features

• Converts PDFs to PNGs and processes images for contour detection.
• Applies OCR for text extraction and outputs results in structured JSON format.
• Detects graph communities using algorithms like Girvan-Newman and Louvain.
• Visualizes results with GraphML files and provides embeddings for clustering.
• Supports advanced debugging and data validation steps.

Installation

Prerequisites

• Python 3.x
• Virtual environment setup
• Jupyter Notebook installed
• Required Python libraries: pdf2image, opencv-python-headless, numpy, pandas, and PaddleOCR

Step-by-Step Installation

Follow the setup instructions provided in the Structure.pdf document or refer to the "Installation Guide" section of the repository.

Quick Setup

1. Create and activate a virtual environment:

   python3 -m venv accimap_env
   source accimap_env/bin/activate  # macOS/Linux
   accimap_env\Scripts\activate     # Windows

2. Install dependencies:

   pip install -r requirements.txt

3. Launch the Jupyter Notebook:

   jupyter notebook

Usage

1. Open the PDF2_Multiple.ipynb notebook in Jupyter.
2. Follow the step-by-step execution of the cells as described in Installation Guide.pdf.
3. View results in structured JSON formats, debug images, or GraphML files for analysis.

Output

• Processed Files: Generated PNGs, debug images, and JSON outputs for text and rectangle data.
• Graph Analysis: Adjacency matrices, GraphML files, and clustering results.
• Visualisation: Community graphs with modularity and embedding-based clustering.

Documentation

For detailed documentation, refer to:

• Structure.pdf for setup guidance.
• Installation Guide.pdf for detailed workflow and cell descriptions.

Contributing

Contributions are welcome! Please fork the repository and submit a pull request for any enhancements or bug fixes.

Licence

MIT Licence