The effect of temporal sampling intervals on typical human mobility indicators obtained from mobile phone location data

Mobile phone location data have been extensively used to understand human mobility patterns through the employment of mobility indicators. The temporal sampling interval (TSI), which is measured by the temporal interval between consecutive records, determines how well such data can describe human activities and influence the values of human mobility indicators. However, systematic investigations of how the TSI affects human mobility indicators remain scarce, and characterizing those relationships is a fundamental research question for many related studies. This study uses a mobile phone location dataset containing 19,370 intensively sampled individual trajectories (TSI < 5 minutes) to systematically assess the impacts of the TSI on four typical mobility indicators that describe human mobility patterns from different aspects, which are movement entropy, radius of gyration, eccentricity, and daily travel frequency. We find that different TSIs have complex impacts on the values of different mobility indicators. Specifically, (1) coarser TSIs tend to underestimate the values of the four selected indicators with different degrees; (2) the degrees of underestimation vary significantly among users for eccentricity and daily travel frequency but exhibit high inter-user consistency for radius of gyration and movement entropy. The above findings can help better understand the variations among human mobility studies.

移动手机位置数据已被广泛借助移动性指标，用于解析人类移动模式。时间采样间隔（Temporal Sampling Interval, TSI）指连续记录之间的时间间隔，其不仅决定了此类数据刻画人类活动的精度，还会对移动性指标的数值产生影响。然而，目前针对TSI如何影响移动性指标的系统性研究仍较为匮乏，厘清二者间的关联正是诸多相关研究的核心科学问题。本研究采用一份包含19370条高密度采样个体轨迹的手机位置数据集（TSI<5分钟），系统评估了TSI对四类从不同维度刻画人类移动模式的典型移动性指标的影响，这四类指标分别为移动熵（movement entropy）、回转半径（radius of gyration）、偏心率（eccentricity）以及每日出行频次（daily travel frequency）。研究发现，不同TSI对不同移动性指标的数值存在复杂影响。具体而言：（1）采样间隔越粗（即采样频率越低），会以不同程度低估上述四类选中指标的数值；（2）对于偏心率和每日出行频次，不同用户间的低估程度存在显著差异；而对于回转半径与移动熵，不同用户间的低估程度则表现出较高的一致性。上述研究结果有助于更深入地理解不同人类移动研究间的差异。