基于时间重定标的人类行为时间特性成因研究

Research on Uncovering Origin of Temporal Activity Based on Redefined Interevent Time

随着大数据技术的发展,人类社会活动积累了海量的数据。通过这些数据,对人类活动的时空特性进行分析发现人类活动的时间间隔表现出近似幂律分布的特点,并且人类活动的幂律指数随用户活跃性的提高而增长。为了研究这种幂律现象的成因,提出了一种改进的时间重定标算法,选取个体两个相继行为发生的时间间隔内,同一时间其他个体所发生的行为总数与该个体平均时间间隔的乘积作为新的时间度量。新的算法可以消除用户活跃性的周期和波动对实验结果的影响,并且兼顾个体活跃性的作用。最后以该度量重新对个体行为的时间间隔分布进行了分析。实验结果表明,在新的时间度量下,其仍然表现出幂律的分布特点。说明人类活动的幂律特性的成因与用户活跃性的周期和波动无关,而是由个体内禀的特性引起。

With the development of big data technology,human social activities have accumulated a huge amount of data.According to these data,the analysis of the temporal and spatial characteristics of human activities reveals that the time interval of human activities exhibits the characteristics of approximate power law distribution,and the power law index of human activities increases with the increase of user activity.In order to study the cause of this power law phenomenon,an improved time rescaling algorithm is proposed.The total number of behaviors of other individuals between the time interval in the occurrence of two successive behaviors of individuals,which multiplied by the average time interval of the individual,is taken as the new time measure.The new algorithm can eliminate the impact of user activity cycle and fluctuation on the experimental results,and take into account the role of individual activity.Finally,the time interval distribution of individual behavior is analyzed again with this metric.The experiment shows that under the new time metric,it still shows the distribution characteristics of power law,which means that the cause of the power law characteristic of human activity is not related to the period and fluctuation of user activity,but is caused by the inherent characteristics of individuals.