Javascript must be enabled for the correct page display

Identifying relationships between websites, relating websites to people

Hoekstra, M (2015) Identifying relationships between websites, relating websites to people. Master's Thesis / Essay, Computing Science.

[img]
Preview
Text
thesis.pdf - Published Version

Download (2MB) | Preview
[img] Text
toestemming.pdf - Other
Restricted to Backend only

Download (514kB)

Abstract

Finding ownership relations between websites in the almost 1 billion websites available on the Internet is like searching for a needle in a haystack. Although website owners have the possibility of estab- lishing connections with their other websites, it might not always be desired in order to remain anonymous. Even if these connections are established this is not done following a uniform protocol, making the identification of such relations an intricate task. Nevertheless, identifying these relations yields a broad range of opportunities. One can utilize this information in law enforcement, commerce or use it for optimizing search engines. In this work I present my investigation on whether it is possible to identify relations between websites using their characteristics. These characteristics consist of identifiers, which are unique across websites, as well as detectable website technologies (e. g. servers, frameworks) serving as features. Logistic regression is used with the features as in- put and identifiers at the base of the ground truth. Relations between websites are identified with an accuracy of up to 90%, measured with an F1 score (i. e. harmonic mean of precision and recall).

Item Type: Thesis (Master's Thesis / Essay)
Degree programme: Computing Science
Thesis type: Master's Thesis / Essay
Language: English
Date Deposited: 15 Feb 2018 08:09
Last Modified: 15 Feb 2018 08:09
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/13412

Actions (login required)

View Item View Item