Institutional Repository of Vilnius University
Deploying and analyzing honeypot for cybersecurity research /
Abstract
dc:descriptionThe increasing number of cyberattacks targeting SSH and HTTP services puts a light on the need for effective cybersecurity education. However, existing deception tools for honeypots, such as Cowrie or T-Pot, are often overly complex and resource intensive, making it difficult for beginners to configure and utilize them effectively in the learning process. The aim of this thesis is to develop a lightweight, containerized honeypot system specifically designed for cybersecurity education that simplifies the analysis of SSH and HTTP threats through AI driven log interpretation. The thesis addresses the gap in cybersecurity education domain where existing tools are often too complex or resource intensive for students and novices. The objectives were to analyze market trends and existing solutions, design the system architecture, develop the logging pipeline, and evaluate the system's detection capabilities through simulated attacks. The system was developed using tools including Python (Paramiko and Flask) for backend emulation, React.js for the user interface, and the Elastic Stack (Filebeat, Elasticsearch) for data management. Docker was utilized to ensure safe, isolated deployment. A key feature of the development is the integration of the Google Gemini API to provide Artificial Intelligence (AI) driven log summarization and dynamic filesystem generation. The result of the thesis is the creation of “Pooh’s Honeypot”, an educational tool that captures unauthorized access attempts and command executions in real time. Unlike traditional tools, it translates complex raw logs into natural language explanations. The conclusions indicate that the developed system successfully balances realism with safety and usability. Testing confirmed that the honeypot captures 100% of executed credentials and commands during simulations. The integration of AI significantly reduces the cognitive load for beginners, making it an effective tool for learning fundamental cybersecurity concepts compared to more complex enterprise solutions like T-Pot or Cowrie.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Muslim, Muhammad,
Subjects
dc:subject × 1Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- eng
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD277690929&prefLang=en_US