Twitter Topic Evolution/Information Diffusion Research

Paper Reads and Summary 

Y. Yang

21/12/2015

Agenda

• An Analysis of Topical Proximity in the Twitter Social Graph

• Predicting the Speed, Scale, and Range of Information Diffusion in Twitter

• Differences in the Mechanics of Information Diffusion Across Topics: Idioms, Political Hashtags, and Complex Contagion on Twitter 

An Analysis of Topical Proximity in the Twitter Social Graph

University College Dublin, UCSB

The relationship between content similarity and social distance in the social network

Investigated the utility of prominent features such as Retweets and Hashtags as predictors of similarity

We aim to test whether and to which extent distance in a social network correlates with content similarity. 

What is good

assess two types of signals per user independently, with respect to both of their roles as target and source of feedback respectively, while processing feedback events in their temporal order, not only the quality of the user as content creator is evaluated, but also the quality of the user as a content evaluator

real-world user relationships has been highlighted as an important factor to leverage improved recommendation quality. For instance, modeling the reputation of users to bias future recommendations from users who are both relevant and reputable

What is good