Item Details

title

Measurements Associated With Learning More Secure Computer Configuration Parameters

author

Zhou, Xin

abstract

To defend against a cyber attack in which the attacker searches the network for vulnerable machines, most people will install some specific security software that cost some money, or download the newest patches for the softwares to avoid some vulnerabilities. Neither way can be efficient especially when attacks are always updated. Reconnaissance is the essential part of a cyber attack during which the attacker is to learn about vulnerabilities of the targeted machines, including credentials, software versions, and misconfigured settings. Moving target strategy then can be implemented specifically against this part of an attack. Theoretically, a change in configuration during the construction of an exploit will alter the computer such that the machine no longer contains the same vulnerabilities discovered during reconnaissance, thereby rendering the initial reconnaissance step ineffective.

subject

Genetic Alogrithm

Machine Learning

Moving Target

Resilience

contributor

John, David J (committee chair)

Fulp, Errin (committee member)

Gage, Don (committee member)

Turkett, William (committee member)

date

2015-06-23T08:35:48Z (accessioned)

2016-06-22T08:30:09Z (available)

2015 (issued)

degree

Computer Science (discipline)

embargo

2016-06-22 (terms)

identifier

http://hdl.handle.net/10339/57131 (uri)

language

en (iso)

publisher

Wake Forest University

type

Thesis