I research, design and code technologies that can improve how we perceive and produce information. I design incentives to encourage positive user behavior through the user interface and diverse modalities to communicate information. My work focuses on news reporting, better engagement with information from news and search results, crowdsourcing journalism and remote reporting.

I am visiting Microsoft Research Fuse Labs to work on technologies for local communities, crowdsourcing local news and supporting local bloggers.

I finished my Master in Computer Science with a focus on Human Computer Interaction at the School of Engineering and Applied Science at Harvard. My past research focused on designing interfaces in information retrieval to encourage better queries and discovery of search results. I researched how users perceive opinions online. I used different ways of organizing information, analyzing user affect, and online user profiles to analyze how information is perceived. I also experimented with crowdsourcing, social network analysis, persuasive technologies, mobile computing and mining big data from the blogosphere.

Visiting Researcher at MSR Fuse Labs eagapie@post.harvard.edu


Feb 5th '14 Invited talk at the University of Washington dub seminar

Jan '14 I am part of the Program Comittee for CHI'14 Work in Progress

Oct '13 SXSW 2014 panel accepted on The LOLs of Nations: Understanding Global Memes

Sept '13 I began a half year visit to Microsoft Research FUSE Labs to work on hyperlocal social research. See my invited talk this summer here.

May '13 This summer I will be interning at JPL/NASA working on interactive data visualization tools.

Apr'13 SIGIR 2013 paper accepted: "Looking Ahead: Query Preview in Exploratory Search", Pernilla Qvarfordt, Gene Golovchinsky, Tony Dunnigan, Elena Agapie

Feb'13 CHI 2013 paper: "Leading People to Longer Queries", Elena Agapie, Gene Golovchinsky, Pernilla Qvarfordt receives Best Paper Honorable Mention Award (top 5% or submissions).

Recent Projects

Local Community News Creation , Microsoft Research, FUSE Labs, Fall 2013-to date

We explore a novel approach for producing news reports of geographic-specific events using locative crowdsourcing. We worked with local bloggers to identify people's needs and abilities to produce local stories. This work explores: the identification of basic templates for event reports, the task decomposition involved in creating them, the design of a system that automates the coordination of on-demand workforce to execute and produce said event reports. We evaluate the system by reporting on several events in different U.S. metropolitan areas.

  • Initial phases of the project here.

News Sharing Through Memes , MIT Media Lab, News and Participatory Media, 2013

This research analyzes how images are used to describe events. It particularly focuses on the memes that get generated throughout an event, the sources they come from and the threads of discussion they reflect.

Persuasive Technologies for Search Interfaces , Fuji Xerox Palo Alto Research Lab (FXPAL), Summer 2012

We designed interfaces for search that encourage people to type longer queries and increase the diversity of documents seen in a search session. Results suggest that our interface manipulations may be effective for eliciting longer queries, but the effect is compromised when instructions to create longer queries are given. Our work also shows how the design of a search preview widget can lead users to engage with more diverse documents.

Communicating Model Based Systems Engineering Information through Data Visualization, Jet Propulsion Lab, NASA, Summer 2013

In the realm of systems engineering, data is organized across a multitude of tables that rank components in all kinds of dimensions. These tables are complicated enough as individual hierarchies of information and need to communicate crucial dependecies that span across them. In a design team of 3 people we did user research, prototyped and built an interactive data visualization tool. This tool supports engineers with different needs and background in assesing rapidly what changes in the system, to get cross system awareness and to explore the data.