The project aims at using advanced machine learning methods on graphs and apply them to scrapped dark-web data, so as to predict future crimes. This would help in further analysis of the commited crimes.
We have primarily used Node2vec whose brief and precise explaination has been provided in the this presenatation.
- Developed a secure pen‑drive mounted connection to the Dark Web using I2P and TOR. Tails OS was the primary OS used.
- Scripted an automated web spider using Python‑Selenium wrapper, that snowballed between various Dark‑web directories to extract useful data, using CRISP‑DM approach
The repository that contains the code for the same: Jash-2000/Dark-Web-Spiders