Compare Measuring User Influence in Twitter and Influentials, Networks, and Public Opinion Formation

Two Papers

1. Measuring_User_Influence_in_Twitter:_The_Million_Follower_Fallacy, Cha, M. and Haddadi, H. and Benevenuto, F. and Gummadi, K.P., ICWSM 2010
2. Influentials,_Networks,_and_Public_Opinion_Formation, Watts, Duncan J., and Peter Sheridan Dodds. Journal of consumer research 34.4 (2007): 441-458.

Problem

Both paper address the same problem i.e. understanding the process of public opinion formation. Watts's paper establishes a theoretical foundation for information diffusion and Cha's paper offers the empirical evidence on a real-world social network: Twitter.

Dataset

Watts's paper only uses synthetic dataset. However, Cha's paper adopt the Twitter dataset consists of 54M users and 1,755,925,520 tweets.

Big Idea

The idea of Watts's paper is to first propose a theoretical model and simulate the model on synthetic data from which the conclusions are deduced. The Cha's paper reverses Watt's process and it starts from collecting real-world dataset from which the conclusions are induced. Since the two work are complementary, their common conclusions are quite convincing.

Evaluation Criteria

Watts's paper uses the size of the cascade in the influence network as the primary criteria for evaluating influence. However, Cha's paper measures the influence by Spearman’s rank correlation coefficient in which they first compute the influence rank lists under different measurement (in-degree, #retweets and #mentions) and then calculate the pari-wise correlation coefficient.

Additional Questions

• How much time did you spend reading the (new, non-wikified) paper you summarized? 3.5 hour.
• How much time did you spend reading the old wikified paper? 30 minutes.
• How much time did you spend reading the summary of the old paper? 15 minutes
• How much time did you spend reading background materiel? 30 minutes
• Was there a study plan for the old paper? No