Carnegie Mellon University
Browse
yuanyil_master_heinz_2024 (1).pdf (269.85 kB)

Data Poisoning in Network Anomaly Detection Systems

Download (269.85 kB)
thesis
posted on 2024-06-25, 16:46 authored by Yuanyi LiYuanyi Li

We investigate the performance of the isolation forest anomaly detection algorithm under data poisoning. We design an experiment using an empirical cybersecurity dataset called BETH, and we report model performance as the training data are incrementally poisoned. We find that, while it may be feasible for attackers to use data poisoning to prevent an anomaly detection model from alerting to their attack, we also find that the isolation forest contains some robustness against this style of attack. Finally, we acknowledge the limitations of our experiment and provide recommendations for future research 

History

Date

2024-05-03

Degree Type

  • Master's Thesis

Department

  • Heinz College of Information Systems and Public Policy

Degree Name

  • Master of Science in Information Security Policy and Management (MSISPM)

Advisor(s)

Clarence Worrell

Usage metrics

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC