By MortazaviBlog on May 29, 2006
Thanks to my colleague and friend Sivakumar Thyagarajan for pointing me to the 3rd Annual Workshop on the Weblogging Ecosystem (hosted by blogpulse) and to the paper by Kyumars Sheykh Esmaili, Mohsen Jamali, Mahmood Neshati, Hassan Abolhassani and Yasaman Soltan-Zadeh, "Experiments on Persian Weblogs."
This paper, authored by researchers at the Semantic Web Reserach Laboratory at Sharif University of Technology focuses on an analysis of the blogs created in the "persianblog.com" domain.
What I found interesting in the paper may perhaps be mundane to many of you. The authors suggest that if the ratio of internal links of a group of blogs to external links grows beyond a certain threshold (2? 2.5?), the group of blogs can be considered to form a separate social network.
The authors suggest that the slow rate of convergence for the Page-Brin "page-rank" equation could be due to internal, small-sized communities within a larger social network.
In comparison, they note the rapid convergence of the Kleinberg "authority-hub" equation (when it is calculated devoid of a query, which is not the common way of doing it) to a result that has little similarity to "page-rank".