Characteristics of human mobility patterns revealed by high-frequency cell-phone position data

Abstract Human mobility is an important characteristic of human behavior, but since tracking personalized position to high temporal and spatial resolution difficult, most studies on patterns rely sparsely sampled data. In this work, we re-examined via comprehensive cell-phone data recorded at a frequency up every second. We constructed networks found that individuals exhibit origin-dependent, path-preferential in their short time-scale mobility. These behaviors are prominent when the high, thus overlooked previous studies. Incorporating measured quantities from our into conventional models shows inconsistent statistical results. finally revealed individual preferential transition mechanism characterized by first-order Markov process can quantitatively reproduce observed travel both population levels all relevant time-scales.