ORCID Identifier(s)

0000-0003-0146-466X

Graduation Semester and Year

2019

Language

English

Document Type

Thesis

Degree Name

Master of Science in Computer Science

Department

Computer Science and Engineering

First Advisor

Shirin Nilizadeh

Abstract

The public at large is increasingly concerned with privacy online. While the focus is on the data privately collected by platforms, there are also privacy concerns in the realm of public data. Seemingly innocuous information shared in public, on online platforms, can be pieced together to detrimentally affect one's privacy in unexpected ways. On YouTube there exists a rich public dataset for adversaries to analyze for the purposes of breaching privacy; particularly due to the intersection of location and facial data. The goal of this work is to characterize the privacy risks that exists on YouTube, and explore the viability of large-scale analysis of video data through easily accessible means for the purposes of identifying users by face in different videos. This work's threat model presumes an adversarial actor is interested in identifying faces across several videos. In phase one focus was on the efficient collection of visual data. In phase two the compiled dataset was characterized, and it's associated facial images were analyzed with Microsoft Azure Face API. In phase three the data obtained in the prior steps was analyzed with a focus on the implications to personal privacy. In conclusion, this work finds that there may be privacy concerns for bystanders who are unaware they are being recorded in public, and for video publishers who are relying upon security through obscurity. Specifically, it was possible to identify the same persons across several videos and YouTube channels in the San Francisco area within a two-week time span.

Keywords

Privacy, Face recognition, Facial recognition, Cloud services, Cloud computing, Social media, Large scale data analysis

Disciplines

Computer Sciences | Physical Sciences and Mathematics

Comments

Degree granted by The University of Texas at Arlington

28858-2.zip (7434 kB)

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.