My scientific research themes are:
- top-K query processing
- human computation
- social media
My Doctoral Major project is about optimizing active and passive crowdsourcing techniques. On one hand, with active crowdsourcing we would like to help algorithms (rather than humans) to solve problems that are too large to be solved for a machine (e.g., analysis of multimedia content). Here, the goal is to minimize the budget required to involve the humans in the loop, while maximizing the quality of the outcome. On the other hand, with passive crowdsourcing we would like to automatically analyze the content that users post on social media every day, and infer some statistics from it (e.g., who are the influencers on a specific topic).
My Doctoral Minor project is about inferring phylogenies from videos posted on social media such as YouTube. Users are daily posting videos that are either original (i.e., produced by themselves) or copied from other users. Still, copying videos from others has its own drawbacks: copyright infringement, duplication of content over the web. In this work, which is still about passive crowdsourcing, we try to infer the ancestry relationships between videos, at first finding the duplicates and then connecting them with directed edges. As a result, identifying the roots of the produced graph corresponds to identify the original videos.