Scale Behavior Research In Human Correspondence Patterns

In the early days, due to the lack of statistical tools and methods, alarge amount of data were simplified, so people roughly thought thatoccurring time of people behavior was uniform. The silence of long timeand the outbreak of a short period of time was ignored, thus humanbehavior were depicted by poisson process, namely, inter-event timebetween two consecutive events follows negative exponential distribution.While, more and more results show that it is not the case. Correspondencebehavior is an important role in social relations, so statistical analysis ofthe correspondence database becomes one of the most importantapproaches to reveal the social structure and it has been payed widelyattention.The main contributions are summarized as follows. 1. Review the development history of human dynamics for more than tenyears, including the present situation, bottle-necks and existed problemsin national and overseas research. Also discuss the principle andconclusions for respective model specifically.2. In order to understanding of the new relevant research dynamic, thisarticle summarizes the latest research results on communication of thehuman dynamics in recent years, including written communication, email,phone, short-message, BBS, MSN, QQ, blog and so on.3. In this paper, through the statistics from Qian Xuesen correspondence,we present that response size distribution of the letters is a exponentdistribution with the best exponent0.005, implying that humancorrespondence pattern differs from the human interactive behavior inWorld Wide Web.4. we show the underlying driven to correspondence activities byrescaling and collapsing the response time distributions for Qian Xunsencorrespondences. It is possible to characterize some aspects of humancorrespondence mode. Indeed, its practical value and economic value stillencourage scientists to continue exploration.