人类行为动力学统计特性研究——基于在线电影点播实证数据  

Study of the Statistical Properties of Human Behavior Dynamics——Based on the Database of the Online Movies on Demand and the Film Score

作　　者：刘红玲[1] 邓明森 任汉卿[1] 郝进欣[1] 

机构地区：[1]贵州师范学院物理与电子科学学院,贵州贵阳550018 [2]贵州省纳米材料模拟与计算重点实验室,贵州贵阳550018

出　　处：《贵州大学学报（自然科学版）》2014年第4期137-140,共4页Journal of Guizhou University:Natural Sciences

基　　金：贵州省科学技术基金项目(黔科合J字[2011]2113号)

摘　　要：本文基于在线电影点播的实证数据库进行研究,根据电影评分将电影划分为不同的集群,对集群内人们电影点播的时间特性进行统计力学研究,发现偏离泊松分布而用幂函数能够很好的拟合,并进一步对电影的点播率进行统计分析,发现存在任务驱动和兴趣驱动的人类行为模式,电影点播现象存在马太效应。In the context of the age of big data,along with the continuous development of science and technology,electronic commerce gradually expands into human activities,which provides a large number of databases for empirical research of human dynamics.based on the Research of a large number of empirical database of the online movies on demand and the film score,this paper will divide the film into different clusters,and studies the time characteristics of clusters on movie demanded from the perspective of statistical mechanics,finding that it deviates from the Poisson distribution but conforms to the power-law distribution,and the further analysis of the calculated statistics of film demanded rate finds the models of human tusk and interest driven behavior,and the online movies on demand phenomenon exists Matthew effect.

关 键 词：人类行为动力学 电影点播率 时间间隔 幂律分布 

分 类 号：C812[社会学—统计学]