Agregátor stránek temného webu s automatickou detekcí a kategorizací obsahu

Abstract

This thesis focuses on the design and implementation of a system for automated monitoring and categorization of dark web pages. The primary goal is to develop an efficient web crawler capable of traversing and analyzing hidden content within the TOR network. The system enables not only continuous monitoring of website availability but also categorization based on content and extraction of key artifacts such as cryptocurrency addresses.

Content categorization is performed using a language model that automatically classifies pages into thematic groups. All collected information is stored in a database and visualized through a web interface, allowing efficient search and analysis. The results of the monitoring and analysis are described in detail in this thesis, including an overview of the discovered pages, content classification, and identified cryptocurrency addresses.