Analysis of Blog graph by CMU researchers : Interesting observations

Temporal patterns: For the two months of observation,we found that blog posts do not have a bursty behavior;they only have a weekly periodicity. Most surprisingly,the popularity of posts drops with a power law, instead of exponentially, that one may have expected. Surprisingly, the exponent of the power law is -1.5, agreeing very well with Barabasi’s theory of heavy tails in human behavior .

Patterns in the shapes and sizes of cascades and blogs: Almost every metric we measured, followed a power law. The most striking result is that the size distribution of cascades (= number of involved posts), follows a perfect Zipfian distribution, that is, a power law with slope =-2. The other striking discovery was on the shape of cascades. The most popular shapes were the “stars”, that is, a single post with several in-links, but none of the citing posts are themselves cited.

Generating Model: Finally, we design a flu-like epidemiological model. Despite its simplicity, it generates cascades that match several of the above power-law properties of real cascades. This model could be useful for link prediction, link-spam detection, and “what-if” scenarios

Link : www.cs.cmu.edu/~jure/pubs/blogs-sdm07.pdf