Modeling information seeking on digital platforms

Through this set of projects we try to analyze large scale data on anonymized search logs and user queries on Q&A platforms to understand how users search for information online to accomplish specific tasks. Re-visiting online search through the lens of search tasks can lead to the design of task-aware platforms which can offer better personalization, ad targeting and task-aware ranking of search results.

Relevant projects/papers:
1. Task Preferences across Languages on Community Question Answering Platforms, with Sebastin Santy and Rishabh Mehrotra (Working paper)
2. Deconstructing Complex Search Tasks: A Bayesian Nonparametric Approach for Extracting Sub-tasks, with Rishabh Mehrotra and Emine Yilmaz (Presented at NAACL-HLT 2016)
3. Sessions, Tasks & Topics - Uncovering Behavioral Heterogeneities in Online Search Behavior, with Rishabh Mehrotra and Emine Yilmaz (Presented at SIGIR 2016)
4. Characterizing Users’ Multi-Tasking Behavior in Web Search, with Rishabh Mehrotra and Emine Yilmaz (Presented at CHIIR 2016)