Paper Reading 25: Twit Info

Reference

Authors: Adam Marcus, Michael S. Bernstein, Osama Badar, David R. Karger, Samuel Madden, Robert C. Miller

Affiliation: MIT CSAIL,32 Vassar St., Cambridge MA

Presentation: CHI 2011, May 2011, Vancouver, BC, Canada.

Summary

Hypothesis

This paper proposes twitinfo as a new way to chronologically categorize and view tweets related to major events.

Contents

TwitInfo allows users to browse a large collection of tweets using a timeline-based display that highlights peaks of high tweet activity. An original streaming algorithm automatically discovers these peaks and labels them meaningfully using text from the tweets. Users can drill down to subevents, and explore further via geolocation, sentiment, and popular URLs. They present an algorithm and an user interface as the TwitInfo system.

Methods

To evaluate the user interface, 12 participants , 6 with twitter experience were asked to do basic searches that used the most fundamental and commonly used pieces of the UI.

To evaluate the algorithm, the researchers gathered tweets from three soccer games and one month of earthquakes. For a control for the soccer scenarios, the authors annotated the highlights of the soccer game using the game video and web-based game summaries, without looking at tweets. For a control on the earthquake data, they gathered data from the US Geological Survey on major earthquakes during the time period.

Then participants were given five minutes to understand an event using TwitInfo and five minutes to dictate a news report on the event to the experimenter. Later an interview was conducted to get qualitative data.

Results

When all earthquake related discussion were included, precision rose to 89% and similarly it leads 95% precision in the soccer dataset while all related discussions are inlcuded.

The algorithm has high recall, finding all major earthquakes and most soccer events.

Discussion

This application was a great way of looking at a large data set and finding a way to use the information in a different possibly better way. It was a great example of how human computer interaction can be used outside of just the UI. I thought this was a great project although it could have benefited from a mobile client or if the functionality was actually built into twitter.

http://www.youtube.com/watch?v=yD8mPCbifD4 


http://twitinfo.csail.mit.edu/