Scaling behavior of online human activity

¹ Web Sciences Center, School of Computer Science and Engineering, University of Electronic Science and Technology of China - Chengdu 610054, PRC
² Institute of Computing Technology, Chinese Academy of Sciences - Beijing, 100190, PRC

^(a) shimin.cai81@gmail.com

Received: 14 July 2012
Accepted: 9 November 2012

Abstract

The rapid development of the Internet technology enables humans to explore the web and record the traces of online activities. From the analysis of these large-scale data sets (i.e., traces), we can get insights about the dynamic behavior of human activity. In this letter, the scaling behavior and complexity of human activity in the e-commerce, such as music, books, and movies rating, are comprehensively investigated by using the detrended fluctuation analysis technique and the multiscale entropy method. Firstly, the interevent time series of rating behaviors of these three types of media show similar scaling properties with exponents ranging from 0.53 to 0.58, which implies that the collective behaviors of rating media follow a process embodying self-similarity and long-range correlation. Meanwhile, by dividing the users into three groups based on their activities (i.e., rating per unit time), we find that the scaling exponents of the interevent time series in the three groups are different. Hence, these results suggest that a stronger long-range correlations exist in these collective behaviors. Furthermore, their information complexities vary in the three groups. To explain the differences of the collective behaviors restricted to the three groups, we study the dynamic behavior of human activity at the individual level, and find that the dynamic behaviors of a few users have extremely small scaling exponents associated with long-range anticorrelations. By comparing the interevent time distributions of four representative users, we can find that the bimodal distributions may bring forth the extraordinary scaling behaviors. These results of the analysis of the online human activity in the e-commerce may not only provide insight into its dynamic behaviors but may also be applied to acquire potential economic interest.