Data-Driven Privacy Protection in Voice Personal Assistants


Aidan McPherson


Authors: Aidan McPherson, Song Liao, and Dr. Long Cheng


Faculty Mentor: Dr. Long Cheng


College: College of Computing


ABSTRACT

Voice assistants, such as Google Home and Alexa, have become extremely popular in recent years. The convenience they grant for many people is part of that reason. Voice assistants have many programs that allow them to be as useful and all-encompassing as possible to consumers. A major factor of this is the ability for third parties to create their own programs that one can install on their voice assistant. This, however, has become a cause of alarm for many as to whether data is being taken from these voice assistants by third parties. That is why privacy policies are required for many programs, and this is because privacy policies help address user’s concerns over data protection and can hold a developer or company liable for misuse of that data. However, it is unknown how developed these privacy policies are. This research has led us to investigate the data practices of 15635 privacy policies to analyze their effectiveness in protecting consumer data through analysis of many privacy policies. The results have shown that 6676 privacy policies have no data practices, and this is a major vulnerability for consumers.

Video Introduction

Aidan McPherson 2021 Undergraduate Research Symposium