Cryptography-Deciphering-Jupyter-Notebook

Overview

This repository contains a Jupyter notebook (main.ipynb) implementing cryptographic techniques to decipher substitution cipher messages. Starting with brute-force permutations, it advances to frequency analysis using Moby Dick text, bigram scoring, and culminates in the Metropolis algorithm for probabilistic optimization.

Features

• Permutation Brute-Force: Tests all 27 rotations (A-Z + space) to identify viable decryptions.
• Frequency Analysis: Maps cipher letter frequencies to English norms from Gutenberg texts (Moby Dick); includes interactive swapping for refinement.
• Bigram Score Function: Computes log-probability of letter pairs against reference corpus for English-likeness evaluation.
• Metropolis Algorithm: MCMC optimization swaps letters probabilistically based on score improvements, annealing via T sequence (10.0 to 0.1); deciphers complex messages.
• Examples: Some example messages have been left in, change the message within the variable message.

Setup and Usage

1. Clone repo: git clone https://github.com/MrRedIsBack/CodeBreaking_StatisticalPhysics
2. Open main.ipynb in Jupyter: jupyter notebook main.ipynb
3. Run cells sequentially:
   • Downloads Gutenberg texts (Moby Dick, etc.) automatically.
   • Test permutations on sample messages.
   • Refine via frequency mapping and manual swaps.
   • Apply Metropolis for final decryptions.

License

MIT License – feel free to fork and extend for educational use.