2010 Poster Sessions : Quantifying the Influence of Online Media

Student Name : Jaewon Yang
Advisor : Jure Leskovec
Research Areas: Artificial Intelligence, Information Systems
Mainstream and social media play a crucial role in disseminating information. Although there have been many theories about how online media influence people, few attempts have been made to quantify the influence directly. We develop the methodology to trace short textual phrases (memes) that spread through the Web and develop an influence model that represents the temporal variation in the phrase popularity as a function of the influences of the media sites that mentioned it.

We evaluate our model on a set of 172 million news articles and blog posts collected in 2009. Using our model, we estimate the influence of individual media sites have on the popularity of different topics and phrases. Our analysis shows that the dynamics of phrase popularity around its peak exhibits strong non-linear trends. We also find that the influence drops quickly as time passes, and the amount of influence is significantly different depending on the topic and the type of the media.

Jaewon is a PhD student at Stanford University advised by Prof. Jure Leskovec. His research interests include machine learning, social networks, and graph analysis. He received a BS degree in Electrical Engineering from Seoul National University in Korea. Prior to Stanford, Jaewon worked at an IT start-up in Korea as a software engineer for three years.