Statistical inference from power law distributed web-based social interactions

Power law distributions describe naturally-occurring events such as the magnitude of earthquakes, the diameter of moon craters or the complexity of the nervous system (Albert Lazslo Barabasi & Albert, 1999; Newman, 2005). Many man-made phenomena also tend to exhibit power law distributions, for example, the distribution of book titles sold, the sizes of cities or intensity of wars (Newman, 2005). Social networks, such as co-authorship networks of scientific collaborations or Hollywood actor networks and web-based social spaces, all follow power law distributions (A. L. Barabasi et al., 2002; Ravid & Rafaeli, 2004). This article focuses on the power law distributions found in web-based social spaces and proposes a method to perform statistical inference on data from such distributions in order to explain behavior and social phenomena.