Augmenting Chorus---a crowd-powered conversational system---by enhancing the crowd's collective memory about a conversation with users. We hypothesize that curating the crowd's memory as it relates to a particular topic for a particular user enables that user to have more meaningful multi-session conversations.
Conversing with Data Using Crowds
We are developing a framework for using the crowd as a middle-layer that allows intelligent ML systems to become better at interacting with, and understanding, large datasets.
(N.B.: This was work done when I was a part of the SPQR Lab, from 09/2013 to 10/2015.)
Light-Emitting Data: Inferring Web Browsing Activity From Router LED Blinks
Developed system and algorithms to fingerprint webpages based solely on router LED blink patterns. We propose a modified Edit-distance based algorithm that uses $k$-NN for classification of webpages and evaluate using threat models that reflect real-world uses of privacy-enhancing technologies (e.g., VPN and private windows).
GrayFuzz: Fuzzing Using Side Channels
Developed system and framework to leverage information being leaked in side channels (i.e. power and timing) in order to map out the internal state machine transitions of a device, which allows for more intelligent fuzzing.
Stigmalware: Investigating the Prevalence of Malware in the Clinical Domain
Detection of anomalous signals and malware signatures in medical device network flow traffic and darknet flows. Designed heuristics to filter suspicious traces from connection information, including performing timing analysis and graph analysis.