/WikiGameAI

WikiGame solver using text embedding models

Primary LanguagePython

WikiGame Solver

This project is a solver for the WikiGame, which uses web scraping and text embeddings to find a path between two Wikipedia pages. The solver fetches links on each page and selects the most relevant one based on cosine similarity of the embeddings.

Installation

  1. Clone the repository:

    git clone https://github.com/mitchwolfe1/WikiGameAI.git
cd WikiGameAI

  2. Create a virtual environment and install dependencies:

    python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

Usage

Run the solver with:

python main.py "start_url" "end_url"