Game-theoretic Analysis of Detecting Data Exfiltration

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F16%3A00309463" target="_blank" >RIV/68407700:21230/16:00309463 - isvavai.cz</a>

Result on the web

—

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Game-theoretic Analysis of Detecting Data Exfiltration

Original language description

Exfiltration of important data, such as, databases of clients, credit card numbers, or proprietary data, is part of many dangerous attacks in network security. Administrators may use tools based on machine learning to detect these events. We analyze this problem in the framework of the game theory. We model four scenarios based on two crucial dimensions of the problem: the amount of knowledge about the normal behavior of the computer that exfiltrates the data, and the ability of the attacker to discontinue the normal activity of the computer. We propose algorithms to compute optimal strategies of the players in all four scenarios and compare their structure and effectiveness.

Czech name

—

Czech description

—

Type

V_souhrn - Summary research report

CEP classification

JC - Computer hardware and software

OECD FORD branch

—

Project

—

Continuities

N - Vyzkumna aktivita podporovana z neverejnych zdroju

Publication year

2016

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Number of pages

16

Place of publication

Praha

Publisher/client name

CISCO SYSTEMS (Czech Republic) s.r.o.

Version

—