论文部分内容阅读
定量地理解人类行为是现代科学的中心议题之一,但由于人类行为的复杂性,其规律是难以发掘的.当前使用的人类动力学模型都假设人类行为发生的时间是随机分布的.为更好地解释人类行为中的阵发性与重尾特征,基于习惯的人类动力学模型通过结合考虑排队模型与人类行为特有的习惯特征,根据事件已发生的次数与稳定程度调整分布参数,使用正态分布模拟连续事件发生的间隔时间,并利用随机参数模拟现实中随机出现的打断人类习惯行为的突发事件,进一步通过考虑事件发生的持续时间,以模拟人类对事件持续关注的现象.
Quantitative understanding of human behavior is one of the central topics of modern science, but due to the complexity of human behavior, its laws are hard to find.Human kinetic models currently used assume that human behavior occurs at random times Well-explained human behavior in the paroxysmal and heavy tail features, based on the habit of human dynamics model by combining the characteristics of queuing models and human behavior characteristics, according to the number of events has occurred and the degree of stability to adjust the distribution parameters, the use of positive The state distribution simulates the interval time between consecutive events, and uses random parameters to simulate real-time emergent events that interrupt human habits, and further simulates the phenomenon that human beings continue to pay attention to the events by considering the duration of events.