Anna's Personal Website
Hello, my name is Anna Ablove, and I am a graduate student at the University of Michigan studying Computer Science. For the past year, I have been a part of Professor Roya Ensafi's lab, researching the deployment and circumvention of internet censorship. Additionally, for the past two summers, I have worked as an instructor at Insprit AI, where I taught middle school and high school students to program in Python, apply algorithms to real-world datasets, discuss AI ethics and applications, and build interactive web apps to deploy their machine learning models.
Current and Recent Projects
Click icons to reveal project information
VPNalyzer is an interdisciplinary research project from the University of Michigan that aims to analyze the VPN ecosystem. VPNalyzer consists of three parallel efforts: large-scale quantitative and qualitative user studies, a cross-platform desktop tool for users to test the security and privacy features of their VPN connection, and qualitative studies surveying VPN providers. Currently, we are focusing our attention on the international community using VPNs to bypass censorship, and we are in the process of adapting and extending the questions in our VPNalyzer survey to be more tailored to VPN use for circumvention.
Russia's Sovereign RuNet was designed to build a Russian national firewall. Previous anecdotes and isolated events in the past two years reflected centrally coordinated censorship behaviors across multiple ISPs, suggesting the deployment of "special equipment" in networks, colloquially known as "TSPU". Despite the TSPU comprising a critical part of the technical stack of RuNet, very little is known about its design, its capabilities, or the extent of its deployment. In this paper, presented at IMC 2022, we developed novel techniques and run in-country and remote measurements to discover the how, what, and where of TSPU's interference with users' Internet traffic. I developed traceroute visualizations that illustrated where the deployment of TSPU is located on network paths.
"Operation Homunculi is a stealth/horror game where you must navigate through an eerie prison where all is not as it seems. Let the shadows be your guide and use your powers to avoid, distract, and control the freaks of nature you find lurking in the cells and hallways — but make sure you’re not caught off guard, you can’t fight back!" This game was completed as part of my senior design captone course.
This project explores Sarcasm Detection on Reddit, and focuses on methods analyzing both the context and reply, where the reply is the potentially sarcastic comment and the context is a parent comment or equivalent. We aimed to improve accuracy of sarcasm detection on Reddit by further exploring the role of attention mechanisms, which have been explored in previous literature. We investigated the performances of Hierarchical Attention Networks (HANs) with varying training sizes. We show that a HAN provides surprisingly good results for this task, given a large enough training size.
LinkedIn: Anna Ablove