Publication Details

Final Report - Machine Learning Outlier Detectionin Safetica's Data Loss Prevention System

PLUSKAL Jan. Final Report - Machine Learning Outlier Detectionin Safetica's Data Loss Prevention System. Praha: Safetica Services s.r.o, 2017.
Czech title
Závěrečná zpráva - Detekce anomálií za účelem prevence ztráty dat
Type
summary report
Language
english
Authors
Pluskal Jan, Ing., Ph.D. (DIFS FIT BUT)
Keywords

Machine learning, Outlier detection, Data loss prevention

Abstract

Data loss prevention systems are becoming necessities incorporate computer system deployments. Nowadays, when everything is connected, and BYOD (Bring your own device) methodology is tolerated, even encouraged in many companies, network security administrators are obliged to keep with newest technologies to prevent threats to business resources. Threats might be parts of carefully planned corporate espionage, or simple malware encrypting all resources available to it. No matter which threat, data have to be kept safe and each interaction with critical business resources need to be monitored, authorized and logged for future analysis. In this paper, we discuss state of the art methods used for outlier detection, unsupervised learning, and statistical analysis.

The final report describes designed technical solution, methods that were implemented and their performance.

Published
2017
Pages
16
Publisher
Safetica Services s.r.o
Place
Praha, CZ
Back to top