Early Detection of Data Exfiltration Attacks using Gradient Boosted Decision Trees

The technology sector has been a frequent target of data exfiltration attacks in recent years. While several incidents go unreported, several high-profile cases have become known. Various threat actors have targeted U.S. tech companies, including nation-state hackers, cybercriminals, and hacktivists. (Sabir, et al. 2020). “Gradient Boosted Decision Trees (GBDT) also known as Gradient Boosting Machines (GBMs)“ (Friedman, 2001) can be effectively employed to detect data exfiltration attempts by sequentially adding weak predictive models, it can accurately identify anomalous network traffic patterns that may indicate malicious activity. “GBDTs can provide insights into the importance of unique features in predicting data breaches.” (Friedman, 2001). “Neural networks can be trained to identify unusual patterns in network traffic or user behavior, which may indicate potential data exfiltration attempts.” (Dolhopolov, 2024)

Author

• Balagao, Anthony Agustin

Language

• English

Keyword

• Convolutional Neural Networks
• Data Exfiltration
• Deep Neural Networks
• Machine Learning
• Gradient Boosted Decision Trees
• Deep Learning

Date created

• 2024

Type of Work

• Dissertation

License

• All rights reserved

Rights statement

• In Copyright

GW Unit

• Cybersecurity Analytics

Degree

• D.Engr.

Advisor

• Blackford, J.P.
• Fossaceca, John

Committee Member(s)

• Etemadi, Amir

Persistent URL

•  https://scholarspace.library.gwu.edu/etd/mg74qn062