基于真实数据集的人类移动特性分析

Nowadays, more and more scholars use the methods of statistical analysis and data mining to analyze the collected large-scale human trajectories from the perspective of population level to obtain characteristics of human mobility. Based on existing research, characteristics of human mobility from individual perspective were analyzed. Analysis on characteristics of human mobility based on reality data set was conducted. The Kolmogorov-Smirnov and Anderson-Darling methods were used to examine the goodness of fit. The results show that no matter from population level or individual level, residence time distribution does not completely follow power law, but will have peak value at long period. The frequency distribution of resident visits to each location is relative higher to Zipf distribution. Compared with the tendency of power law distribution decreasing gradually with the increase of travel distance, there is no obvious regularity in travel distance distribution obtained from real data set.